Table 1 Summary of early technical analysis studies published between 1961 and 1987
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1. Donchian (1960) 
Copper futures / Daily 
195960 
Channel 
Not considered 
$51.50 per roundtrip 
The current price was
compared to the two preceding week’s ranges.
This trading rule generated net gains of $3,488 and $1,390, on margin
of $1,000, for a single contract of the December 1959 delivery of copper and
the December 1960 delivery, respectively.

2. Alexander (1961) 
S&P Industrials, Dow
Jones Industrials /
Daily 
18971959, 192959 
Filter (11 rules from 5.0 to 50%) 
Buy & hold 
Not adjusted 
Trading rules with 5, 6, and 8% filters generated
larger gross profits than the B&H (buyandhold) strategy. All the profits were not
likely to be eliminated by
commissions. This led Alexander
to conclude that there were trends in stock market prices. 
3. Houthakker (1961) 
Wheat
and corn futures /
Daily 
192139, 194756 
Stoploss
order (11
rules from 0 to 100%) 
Buy
& hold, Sell & hold 
Not
adjusted 
Most
stoploss orders generated higher profits than the B&H or a sell and hold
strategy. Long transactions indicated
better performance than short transactions. 
4. Cootner (1962) 
45
NYSE stocks /
Weekly 
195660 
Moving
average (1/200
days with and without a 5% band) 
Buy
& hold 
Commissions
of 1% per oneway transaction 
Although net returns from moving average rules were not much
different from those from the B&H strategy, long transactions generated
higher returns than the B&H strategy.
Moreover, the variance of the trading rule was 30% less than that of
the B&H. 
5. Gray & Nielsen (1963) 
Wheat futures / Daily 
192143, 194962 
Stoploss
order (10 rules from 1 to 100%) 
Buy
& hold, Sell & hold 
Not
adjusted 
When applying stoploss order rules to dominant
contracts, there was little evidence of nonrandomness in wheat futures
prices. They argued that Houthakker’s
results were biased because he used remote contracts and that postwar
seasonality of wheat futures prices was induced by government loan programs. 
6. Alexander (1964) 
S&P
Industrials /
Daily 
192861 
Filter, Formula Dazhi, Formala
Dafilt, moving average, and Dowtype formulas 
Buy
& hold 
Commissions
of 2% for each roundtrip 
After
commissions, only the largest filter (45.6%) rule beat the B&H strategy
by a substantial
margin. Most of
the other
trading systems earned higher gross profits than filter rules or the B&H strategy. However, after commissions they could not
beat the B&H. 
7. Smidt (1965a) 
May
soybean futures contracts /
Daily 
195261 
Momentum
oscillator (40 rules) 
Not
considered 
$0.36 per bushel per roundtrip 
About 70% of trading rules
tested generated positive returns after commissions. Moreover, half of trading rules returned
7.5% per year or more. 
8. Fama & Blume (1966) 
30
individual stocks of the DJIA /
Daily 
195662 
Filter (24
rules from 0.5 to 50%) 
Buy
& hold 
0.1% per roundtrip plus other costs 
After
commissions, only 4 of 30 securities had positive average returns per
filter. Even before commissions,
filter rules were inferior to the B&H strategy for all but
two securities. Although
three
small filter rules (0.5, 1.0, and 1.5%) earned higher
gross average
returns (11.4%20.9% per year) per security
when considering only long positions, net returns after transaction costs
were not much different from B&H returns. 
Table 1 continued.
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
9. Levy (1967a) 
200
NYSE stocks /
Weekly 
196065 
Relative
strength (Ratios: 1/4 and 1/26 weeks) 
Geometric
average 
1%
per oneway transaction 
Net
returns of several wellperforming rules were nearly two
or three times
the return of the geometric average, although these rules possessed
slightly higher
standard deviations relative to the geometric average. 
10. Levy (1967b) 
200
NYSE stocks /
Weekly 
196065 
Relative
strength (Ratio: 1/26 weeks) 
Not
considered 
1%
per oneway transaction 
Stocks
having the historically strongest relative strength showed an average price appreciation
of 9.6% over 26 weeks (about 20.1% per year). An annual price appreciation
of all stocks was 12.8%. In general,
stocks that had been both relatively strong and relatively volatile produced
higher profits. 
11. 
9
exchange rates /
Daily 
191929, 195062 
Filter (10 rules from 0.1 to 2%) 
Buy
& hold 
Not
adjusted 
Four
of nine exchange rates had average annual gross returns more than 25% for the
best filter
rules, and
three of them (Belgium, France, and Italy) generated returns above 44%. Filter rules beat the B&H
strategy by large differences in returns. 
12. Van Horne & Parker (1967) 
30
NYSE stocks /
Daily 
196066 
Moving
average (100, 150, and 200 days with 0, 2, 5, 10, and 15% bands) 
Buy
& hold 
Commissions
charged by members of the NYSE 
No
trading rule earned a total closing balance nearly as large as that
generated under the B&H strategy.
Even before transaction costs, gross profits from
each moving average rule were less than that from the B&H. 
13. James (1968) 
232
to 1376 stocks from the CRSP at the /
Monthly 
192660 
Moving
average (7 months = 200 days with 2
and 5% bands) 
Buy
& hold 
Not
adjusted 
Moving average rules could not beat the B&H strategy. The largest average dollar
difference between the moving average rules and the B&H strategy
was very
small. 
14. Van Horne & Parker (1968) 
30
NYSE stocks /
Daily 
196066 
Nonweighted
and exponentially weighted moving averages (200 days with 0, 5, 10, and 15%
bands) 
Buy
& hold 
1%
per oneway transaction 
When applying trading rules to long positions, only 55 of 480 cases (16 different combinations
of rules multiplied by 30 stocks) realized profits greater than those
from the
B&H strategy. For long plus
short positions, a
smaller number of trading rules (36 out of 480 cases) outperformed
the B&H. 
15. Jensen & Benington (1970) 
29
portfolio samples
of 200 NYSE stocks /
Monthly 
193165 
Relative
strength (2
rules from Levy (1967a)) 
Buy
& hold 
Actual
round lot rate 
After
transaction costs, Levy’s trading rules did not perform
better than
the B&H strategy. In
fact,
after explicit adjustment for the level of risk, the trading rules on
average generated net returns less than the riskadjusted B&H
returns. 
Table 1 continued.
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
16. Stevenson & Bear (1970) 
July
corn and soybean futures /
Daily 
195768 
Stoploss
order, filter, and combination of both systems 
Buy
& hold 
0.5
cents per
bushel for both commodities 
For
all systems, a 5% filter rule worked best, which generated larger net
profits or greatly reduced losses relative to the B&H strategy. The
filter rule also
outperformed
B&H for both corn and soybean futures.

17. Dryden (1970a) 
/
Daily 
196267, 196264 
Filter (12 rules from 0.1 to 5%) 
Buy
& hold 
Individual stock: 0.625% per oneway transaction 
Without
transaction costs, filter rules consistently beat the B&H strategy for
both indices and an individual stock.
With transaction costs, the returns from the best filter rules were
similar to those from the B&H, but long transactions beat the B&H. 
18. Dryden (1970b) 
15
/
Daily 
196364, 196667 
Filter (14 rules from 0.2 to 6%) 
Buy
& hold 
Not
adjusted 
There
was considerable variation among individual stocks’ returns. On average, filter returns were less than
the corresponding B&H returns except for two smallest filter
rules. However, returns only from long transactions were much higher than the B&H
returns. 
19. Levy (1971) 
548
NYSE stocks /
Daily 
196469 
32
forms of a fivepoint chart pattern 
Buy
& hold 
2%
per roundtrip 
After
transaction costs, none of the 32 patterns for any holding period generated
profits greater than average purchase or shortsale opportunities. Even the bestperforming pattern produced
adjusted relativetomarket returns of 1.1% and 0.1% for oneweek and
4week holding periods, respectively.

20. Leuthold (1972) 
30
live cattle futures contracts /
Daily 
196570 
Filter (1, 2, 3, 4, 5, and 10%) 
Not
considered 
Commissions
of $36 per roundtrip 
Four
of six filters were profitable after transaction costs. In particular, a 3% filter rule generated an
annual net
return of 115.8% during the sample period. 
21. Martell & Philippatos (1974) 
September
wheat and September soybean futures contracts /
Daily 
195669 (195870)^{*} 
Adaptive filter model and pure information model 
Buy
& hold / Optimized trading rules 
Adjusted
but not specified 
As an optimal filter size for period t, the adaptive
model utilizes a filter size which has yielded the highest profits in t1,
subject to some minimum value of the average relative information gain. The pure information model chooses as an
optimal filter size in period t the one with the highest relative average
information gain in period t1. Both models yielded higher net returns than
the B&H only for wheat futures.
However, the variance in net profits was consistently smaller than that of the B&H in
both markets. 
^{*} Years in parentheses indicate outofsample periods.
Table 1 continued.
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
22. Praetz (1975) 
/
Daily 
196572 
Filter (24 rules from 0.5 to 25%) 
Buy
& hold 
Not
adjusted 
For 12 of all 21 contracts of 18month length and
all three 8year price series, the B&H strategy showed better performance than
filter rules, with average differences of 0.1% and 2%,
respectively. For the same data set, in 10 of 24 filters the
B&H returns were greater than average filter returns. Thus, filter rules did not seem
to outperform
the B&H strategy consistently. 
23. Martell (1976) 
September
wheat and September soybean futures contracts /
Daily 
195669 (195870)^{*} 
Adaptive filter models and pure information model 
Buy
& hold / Optimized trading rules 
Adjusted
but not specified 
A new adaptive model was developed and applied to
the same data set as that used in Martell and Philippatos (1974). The new model selects its optimal filter
size for next period based on profitability (e.g., the highest cumulative net
profits) and information gain.
Although the model outperformed the previous adaptive model for around
80% of the sample period, it neither indicated any stability with respect to
the information constraint nor beat the pure information model that allows a
filter size in a particular period to reflect new information. 
24. Akemann & Keller (1977) 
Industry
groups from S&P 500 Stock Index /
Weekly 
196775 
Relative
strength 
S&P
500 Index 
2%
per roundtrip 
The
relative strength rule is designed to buy the strongest stock group in a given
thirteenweek period and sell it after 52 weeks. After adjustment for transaction costs, the
mean return
differential between all 378 possible trials and the market index appeared to
be 14.6%, although the differentials were quite volatile. 
25. Logue & Sweeney (1977) 
Franc/dollar spot exchange rate /
Daily 
197074 
Filter
(14 rules from 0.7 to 5%) 
Buy
& hold 
0.06%
per oneway transaction 
Most
trading rules (13 out of 14 rules) outperformed the B&H strategy after considering
transaction
costs. Compared to the buy
and hold and invest in French government securities strategy, only four filters
failed to generate higher profits. 
26. Cornell & Dietrich (1978) 
6
spot foreign currencies (mark, pound, yen, Canadian dollar, Swiss
franc, and Dutch guilder) /
Daily 
197375 
Filter
(13 rules from 0.1 to 5%), and moving average (10, 25, and 50 days with 0.1 to 2%
bands) 
Buy
& hold 
Computed
by using the average bidask spread for all trades. 
For
the Dutch
guilder, German mark, and Swiss franc, the best rules from each trading
system generated over 10% annual net returns.
Although the net returns were relatively small (1% to 4%) for the
British pound,
Canadian dollar, and Japanese yen, they all beat the B&H strategy. Moreover, since none of
the systematic risk (beta) estimates exceeded 0.12, high returns of the three
currencies were less likely to be compensation for bearing systematic risk. 
27. Logue, Sweeney, & Willett (1978) 
7 foreign
exchange rates /
Daily 
197376

Filter
(11 rules from 0.5 to 15%) 
Buy
& hold 
Not
adjusted 
For
every exchange rate (the mark, pound, yen, lira, France franc,
Swiss franc, and Dutch guilder), profits from the best filter rules exceeded those from the B&H strategy
by differences ranging from 9.3% to 32.9%. 
^{*} Years in parentheses indicate outofsample periods.
Table 1 continued.
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
28. Arnott (1979) 
500
stocks from both the S&P 500 Index and the NYSE Composite Index /
Weekly 
196877 
Betamodified
relative strength 
Not
considered 
Not
adjusted 
Regression results indicated that for the base periods of 1 week to 18
weeks, the correlation between the change in (betaadjusted) relative
strength during the base period and that during any subsequent
period was strongly negative. Hence, careless use of relative
strength might lead to serious money loss. 
29. Dale & Workman (1980) 
90day
Tbill
futures at the IMM /
Daily 
197678 
Moving
average (11
rules from 
Not
considered 
$60
per roundtrip 
For
each individual contract, the best trading rules generated positive net
returns, although the rules did not indicate consistent performances over the sample
period. 
30. Bohan (1981) 
87
to 110 S&P industry groups /
Weekly 
196980 
Relative
strength 
Buy & hold on S&P 500 Index 
2%
per year 
There
was a strong correlation between the performance of the strongest and weakest
industry groups in one year and that of the following years,
although the performance of the other groups did not have much predictive
significance. For example, quintile 1
portfolio, which consists of the top 20% of industry groups, generated a
return of 76% higher
than the
B&H on the
market index, while the market outperformed quintile 5 portfolio by 80%. 
31. Solt & Swanson (1981) 
Gold
from London Gold Market and silver from Handy & Harman /
Weekly 
197179 
Filter
(0.5 to 50%) and moving average (26, 52, and 104 weeks with filters) 
Buy
& hold 
1.0% per oneway transaction
plus 0.5% annual fees 
For
gold, a 10% filter rule outperformed the B&H strategy after adjustment
for transaction
costs. However, none of the filter
rules dominated the B&H strategy for either gold or silver. Moving average
rules were not
able to improve the returns for the filter rules as well. 
32. Peterson & Leuthold (1982) 
7
hog futures contracts from CME /
Daily 
197377 
Filter (10 rules from 1 to 10% and
additional 10 rules from $0.5 to $5) 
Zero
mean profit 
Not
adjusted 
All
20 filter rules produced considerable mean gross profits. It seemed that these profit levels exceeded
any reasonable commission charges in most cases. In general, mean gross profits increased with larger
filters, as did variance of profits. 
33. Dooley & Shafer (1983) 
9
foreign currencies in the /
Daily 
197381 
Filter (7 rules from 1 to 25%) 
Not
considered 
Adjusted
but not specified 
Although
results were slightly different for each currency, small filter rules (1, 3,
and 5%) generally produced high profits, while larger filter rules showed
consistent losses. 
Table 1 continued.
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
34. Brush
& Boles (1983) 
168
S&P 500 stocks /
Monthly 
196780, (two data bases were used for
outofsample tests) 
Relative
strength (parameters were optimized on
the development data base over 26 separate 6month test periods) 
Equal weighted 168stock return /
Optimized models 
2%
per roundtrip 
The
top decile annualized excess return of the best model was 7.1% per year over
the equalweighted 168stock return, after adjustment for risk, dividend
yield, and transaction costs. The
model also produced a compounded growth of 15.2% per year after considering dividend yield and
transaction costs, compared to 5.9% for the S&P 500. 
35. Irwin & Uhrig (1984) 
8
commodity futures: corn, cocoa, soybeans, wheat, sugar, copper, live cattle,
and live hogs /
Daily 
196078
(197981)^{*}, 196068 (196972)^{*}, 197378
(197981)^{*} 
Channel, moving averages, momentum oscillator 
Zero
mean profit /
Optimized trading rules 
Doubled
commissions to capture bidask spread (not specified) 
Trading rule profits during insample periods were substantial and
similar across all four trading systems. Outofsample results for optimal trading
rules also
indicated that
during the 197981 period most trading systems were profitable in corn, cocoa, sugar, and soybean
futures markets. The trading rule profits
appeared to be concentrated in the 197381 period. 
36. Neftci & Policano (1984) 
4
futures: copper, gold, soybeans, and Tbills /
Daily 
197580 
Moving
average (25,
50, and 100 days) and slope (trendline) method 
Not
considered 
Not
adjusted 
Trading
signals were incorporated as a dummy variable into a regression equation for
the minimum mean square error prediction.
Then the significance of the dummy variable was evaluated using Ftests. Overall, moving average
rules
indicated some predictive power for Tbills, gold, and soybeans, while the
slope method showed mixed results. 
37. Tomek & Querin (1984) 
3 random price series (each
series consists of 300 prices) generated from corn prices
for each sample period /
Daily 
197580, 197374, 1980 
Moving
average (3/10
and 10/40 days) 
Not
considered 
$50
per roundtrip 
From each of three random prices series, 20 sets of
prices were replicated. The first 20
sets had moderate price variability, the second set large price variability,
and the third set drift in prices.
Both
trading rules failed to generate positive average net profits for all
three groups with an exception of the 10/40 rule for the relatively volatile price group. The results imply that
technical trading rules may earn positive net returns by chance, although they
on average could not generate positive net
profits. 
38. Bird (1985) 
Cash
and forward contracts of copper, lead, tin, and zinc from London Metal
Exchange (LME) /
Daily 
197282 
Filter:
long positions (and cash profits) (25
rules from 1 to 25%) 
Buy
& hold 
1%
per roundtrip 
For
cash and forward (futures) copper, over 2/3 of filter
rules beat the
B&H strategy. Similar results were
obtained for lead and zinc but with weaker evidence. For tin, the results were
inconsistent. Filter rules performed substantially
better in the earlier period (197277). 
^{*} Years in parentheses indicate outofsample periods.
Table 1 continued.
Criteria: Study 
Markets
considered / Frequency of data 
Insample
period 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
39. Brush (1986) 
420
S&P 500 stocks /
Monthly 
196984 
Relative
strength 
Return
of the equal weighted S&P 500 Index 
1%
per roundtrip 
By
avoiding the yearend effect and exploiting beta corrections and the negative
predictive power of onemonth trends, the best model, which was the
generalized least squares beta approach, generated an annual excess return of
more than 5% over the equalweighted S&P 500, after transaction
costs. 
40. Sweeney (1986) 
Dollar/mark and additional 9 exchange rates /
Daily 
197375 (197580)^{*} 
Filter:
long positions (7
rules from 0.5 to 10%) 
Buy
& hold /
Optimized trading rules 
1/8
of 1% of asset value per roundtrip 
Both in and outofsample tests, small filter rules
(0.5% to 5%)
consistently beat the B&H strategy, and transaction costs did not
eliminate the riskadjusted excess returns of filter rules. Eight filter
rules across 6 exchange rates produced statistically
significant excess
returns over
the B&H in
both in and outof sample periods. 
41. 1986) 
/
Daily 
197176
(197781)^{*},
196173
(197481)^{*}, 197478
(197981)^{*} 
A
statistical pricetrend model 
Buy
& hold and interest rate for bank deposit /
Optimized trading rules 
1%
per roundtrip for agricultural futures and 0.2% for currency futures 

42. Thompson & Waller
(1987) 
Coffee and cocoa futures in the NY Coffee, Sugar,
and Cocoa Exchange / 6 weekly sets of transactiontotransaction prices
for each market 
198183 
Filter (for coffee, 5¢ through 35¢ in multiples of 5¢ per 100 lb; for cocoa, $1 through $7 per metric
ton) 
Not considered 
Estimated execution costs 
For both nearby and distant coffee and cocoa
contracts, filter rules generated average profits per trade per contract substantially
lower than estimated execution costs per contract in all cases in which
profits were statistically significantly greater than zero. The estimated execution costs per trade per
contract were $32.25 (nearby) and $69.75 (distant) for coffee futures
contracts and $12.60 (nearby) and $21.80 (distant) for cocoa futures contracts. 
^{*} Years in parentheses indicate outofsample periods.
Table 2 Categories for modern technical
analysis studies
Category 
Number of studies 
Representative study 
Transaction costs 
Risk adjustment 
Criteria Trading rule optimization 
Outofsample
tests

Statistical tests 
Data snooping addressed 
Distinctive features 
Standard 
23 
Lukac, Brorsen, & Irwin (1988) 
√ 
√ 
√ 
√ 
√ 

Conduct parameter optimization and outofsample
tests. 
Modelbased bootstrap 
21 
Brock, Lakonishok, & LeBaron (1992) 

√ 


√ 

Use modelbased bootstrap methods for statistical
tests. No parameter optimization and
outofsample tests conducted. 
Genetic programming 
11 
Allen
&
Karjalainen (1999) 
√ 
√ 
√ 
√ 
√ 
√ 
Use genetic programming techniques to optimize
trading rules. 
Reality Check 
3 
Sullivan, Timmermann, & White (1999) 

√ 
√ 
√ 
√ 
√ 
Use White’s Reality Check Bootstrap methodology for
optimization and statistical tests. 
Chart patterns 
11 
Chang & Osler (1999) 
√ 
√ 


√ 

Use recognition algorithms for chart patterns. 
Nonlinear 
7 
Gençay
(1998a) 
√ 
√ 
√ 
√ 
√ 

Use nearest neighbors and/or feedforward network regressions
to generate trading signals. 
Others 
16 
Neely
(1997) 
√ 
√ 


√ 

Most studies in this category lack trading rule
optimization and outofsample tests, and do not address datasnooping
problems. 
Table 3 Summary
of standard technical analysis studies published between 1988 and 2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1.
Lukac, Brorsen, & Irwin (1988) 
12
futures from various exchanges: agriculturals, metals, currencies, and
interest rates /
Daily 
197583 (197884) 
12
systems (3
channels, 3
moving averages, 3 oscillators, 2
trailing stops, and a combination) 
Zero
mean profit /
Optimized trading rules 
$50
and $100 per roundtrip 
Outofsample results indicated that 4 of 12 systems generated
significant aggregate portfolio net returns and 8 of the 12 commodities earned
statistically significant net returns from more than one trading system. Mark, sugar, and corn markets appeared to
be most profitable during the sample period.
In addition, Jensen test confirmed that the same four trading
systems having
large net returns still produced significant net returns above risk. 
2. Lukac & Brorsen (1989) 
15
futures from various exchanges: agricultural commodities,
metals, currencies, and interest rates / Daily 
196585
(various) 
Channel
and directional movement (both systems had 12 parameters ranging 5
days to 60
days in increments of 5) 
Buy
& hold /
Optimized trading rules 
$100
per roundtrip 
Technical
trading rule profits were measured based on various optimization methods,
which included 10 reoptimization strategies, one random strategy, and
12 fixed parameter strategies. The two
trading systems generated portfolio mean net returns significantly greater
than the B&H strategy. However,
the trading systems yielded similar profits across different optimization
strategies and even different parameters.
Thus, the parameter optimization appeared to have little value. 
3. Sweeney & Surajaras (1989) 
An equallyweighted portfolio and a
variablyweighted portfolio of currencies / Daily 
Prior 250 to 1400day prices (198086) 
Filter, single moving average, double moving
average, and the best system 
Buy
& hold / Optimized trading rules 
Adjusted but not specified 
Most trading systems generated riskadjusted mean
net profits after transaction costs, and the single moving average rule
performed best. The variablyweighted
portfolio approach generally outperformed the equallyweighted approach. Changing neither parameters for each
trading system on a yearly basis nor amounts of data used to select optimal
parameters seem to improve trading profits.

4. Taylor & Tari (1989) 
IMM
currency futures: pound, mark, and Swiss franc; /
Daily 
197478 (197987); (198285) 
A
statistical pricetrend model 
Buy
& hold, Zero mean profit /
Optimized trading rules 
Currency futures: 0.2% per roundtrip; Agricultural
futures: 1% 
During the outofsample period, 197987, the
trading rule earned aggregate mean net return of 4.3% per year for three
currency futures. The mark was the
most profitable contract (5.4% per year).
From 198285, the trading rule generated a mean net return of 4.8% for
cocoa, 4.26% for coffee, and 18.8% for sugar, outperforming the B&H
strategy for cocoa and sugar futures.

5. Lukac & Brorsen (1990) 
30
futures from various exchanges: agriculturals, metals, oils, currencies,
interest rates, and S&P 500 /
Daily 
197585
(197686) 
23
systems (channels, moving averages, oscillators, trailing stops, point and
figure, a countertrend, volatility, and combinations) 
Zero
mean profit /
Optimized trading rules 
$50
and $100 per roundtrip 
Only 3 of 23 trading systems had negative mean monthly portfolio
net returns after transaction costs, and 7 of 23 systems generated net returns significantly
above zero at
10% level. Most of the trading
profits
appeared to be made
over the 197980
period. In the individual commodity markets, currency futures produced the
highest returns, while livestock futures yielded the lowest returns. 
Table 3
continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
6. 
4
currency futures from IMM of the CME: pound, mark, yen, and Swiss franc /
Daily 
197787 (198287) 
3 technical trading systems (filter, channel, moving
average), 2 statistical pricetrend models 
Buy & hold / Optimized trading rules 
0.2% per roundtrip 
All trading rules outperformed the B&H strategy
across all currency futures. Among
trading rules, three technical trading systems and a revised statistical
trend model generated statistically significant and much higher mean net
returns (3.0% to 4.0%) than that (2.0%) of the original pricetrend model for
most currencies. These returns could
not be explained by nonsynchronous trading or timevarying risk premia. 
7. Farrell & Olszewski (1993) 
S&P
500 futures /
Daily 
198290 (198990) 
A
nonlinear trading strategy based on ARMA (1,1) model and 3 trendfollowing
systems (channel and volatility systems) 
Buy
& hold / Optimized trading rules 
0.025%
per roundtrip 
Although
the nonlinear trading strategy were slightly more profitable than the B&H strategy,
the result was statistically insignificant.
For the insample period, the nonlinear optimal
trading strategy was more profitable than the B&H by nearly 5%, while for
the outofsample period, the trading strategy was better by 3%. Meanwhile, the three trend following strategies were
more profitable than the nonlinear trading strategy by
around 5% to
11% during the outofsample period, depending on the trading
strategy. 
8. Silber (1994) 
12
futures markets: foreign currencies, shortterm interest rates, metals,
oil, and S&P 500 /
Daily 
1979 (198091) 
Moving
average (short
averages: 1 day to 15 days; long averages: 16 to 200 days) 
Buy
& hold (& roll over) /
Optimized trading rules 
Bidask
spreads per roundtrip (2 ticks for crude oil and gold; 1 tick for the rest of contracts) 
After
transaction costs, average annual net returns were positive for all contracts
but gold, silver, and the S&P 500.
In particular, most currency futures earned higher
net profits (1.9% to 9.8%). For those profitable
markets,
moving average rules beat the B&H strategy except for 3month
Eurodollars. Test
results using a Sharpe
ratio criterion were similar. Hence,
trading profits appeared to be robust to transaction costs and risk. Central bank intervention is
one of possible
explanations for the trading profits. 
9. 
4
currency futures from IMM: pound, mark, yen, and Swiss franc /
Daily 
1980all
previous contracts (198290) 
Channel 
Zero
mean profits /
Optimized trading rules 
0.2%
per oneway transaction 
For
price series generated by ARIMA(1,1,1) model, channel rules correctly
identified the sign of conditional expected returns with around 60%
probability. During 198290, optimal
channel rules produced an average net
return of 6.9% per year. The ttest
indicated that the return was significant at the 2.5% level. The best trading opportunities occurred for 198587. 
10. Menkhoff & Schlumberger (1995) 
3
spot exchange rates: mark/dollar, mark/yen, and mark/pound /
Daily 
198191, 198185 (198691) 
Oscillator
(33 moving averages) and momentum (10 rules from 
Buy
& hold /
Optimized trading rules 
0.0008
DM for 1$; 0.0017 DM for 1 yen; 0.003 DM for 1
BP per
roundtrip 
During
the outofsample period, 84% out of 129 technical trading rules tested
outperformed
the B&H strategy across exchange rates, after
adjustment for
transaction costs and risk. However,
superiority of optimal trading rules during the insample period deteriorated
in the outofsample period, even though they still outperformed the B&H
strategy. 
Table 3
continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
11. Lee & Mathur (1996a) 
6
European currency spot crossrates /
Daily 
198892
(198993) 
Moving
average (short
moving averages: 1 day to 9 days; long moving averages: 10, 15, 20, 25, and 30
days) 
Zero
mean profits /
Optimized trading rules 
0.1%
per roundtrip 
Results of insample tests indicated that the trading rules did not
yield significantly positive returns for all cross rates but yen/mark and
yen/Swiss franc (11.5% and 8.8% per year, respectively). Outofsample results
were even worse. Most cross rates
earned negative trading returns, although long positions for the
yen/mark produced marginally significant
positive returns. 
12. Lee & Mathur (1996b) 
10
spot crossrates /
Daily 
198892 (198993) 
Moving
average (short
moving averages: 1 day to 9 days; long moving averages: 10, 15, 20, 25, and
30 days) and channel (2 to 50 days) 
Zero
mean profits /
Optimized trading rules 
0.1%
per roundtrip 
During insample periods, moving average rules in general
produced negative
or statistically insignificantly positive net returns except the
mark/yen
(11.5% per year) and the Swiss franc/yen (8.8% per year). Similar results were
found for
channel rules. During outofsample periods, overall returns of the
trading rules
were negative or statistically insignificantly positive. Only for the mark/lira,
both long positions of moving average rules and channel rules generated statistically
significant
profits. 
13. Szakmary & Mathur (1997) 
5 IMM
foreign
currency futures and spots: mark, yen, pound, Swiss
franc, and Canadian dollar /
Daily 
197790 (197891) 
Moving
average (short
moving averages: 1 day to 9 days; long moving averages: 10, 15, 20, 25, and
30 days) 
Zero
mean profits /
Optimized trading rules 
0.1%
per roundtrip 
Insample
results indicated that moving average rules generated both statistically and
economically significant returns for all currency futures but the Canadian
dollar. Similar results were reported
for both outofsample data (annual net returns ranged from 5.5% to 9.6%) and spot rates. Further analyses showed that the moving
average rule profits resulted from the central bank’s “leaning against the
wind intervention.” 
14. Goodacre, Bosher,
& Dove (1999) 
254 companies in the FTSE 350 Index and 64 option
trades in the / Daily 
Prior 200 days (198896) 
CRISMA
(combination
system of Cumulative volume, RelatIve Strength,
and Moving Average) 
FTSE All Share Index / Optimized parameters 
0
to 2% per roundtrip 
The CRISMA trading system generated annualized
profits ranging 6.9% to 19.3% depending on transaction costs, while an
annualized return on the FTSE All Share Index over the same time period was
14.0%. When adjusted for market movements
and risk, however, mean excess returns for nonzero levels of transaction
costs were significantly negative.
Moreover, performance of the trading system was not stable over
time. With option trading, the system
generated mean return of 10.2% per trade even in the presence of maximum
retail costs, but only 55% of trades were profitable. 
15. Kwan, Lam, So,
& Yu (2000) 
Hang
Seng Index Futures /
Daily 
198697
(199098) 
A
statistical pricetrend model 
Buy
& hold / Optimized
parameters 
0.4
to 0.5% per oneway transaction 
The
pricetrend model performed poorer than the B&H strategy in the periods
199193 and 199596 when the market was bullish. However, the trading rule produced larger
profits than the B&H in the years, 90, 94, 97, and 98 when the market
became up and down. Across all years
and transaction
costs considered, an average net return (10.1%) of the trading rule
was slightly smaller than that (13.5%) of the B&H strategy. 
Table 3
continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
16. Maillet & Michel (2000) 
12
exchange rates (combinations of U.S. dollar, mark, yen, pound, and /
Daily 
197479 (197996) 
Moving
average (short
moving averages: 1 day to 14 days; long moving averages: 15 to 200 days) 
Zero
mean profits, buy & hold /
Optimized trading rules 
Not
adjusted 
Optimized moving average rules generated statistically significant
returns and outperformed the corresponding B&H strategies with the
exception of the mark/franc rate. Bootstrap tests generally
confirmed the results with the rejection of higher returns only
in 4 out of 12
rates: the mark/dollar, mark/franc, yen/dollar, and yen/franc. Moreover, riskiness of both moving
average rules
and the B&H strategy, which was measured by their standard deviations, appeared
to be not much different. 
17. 
1)
Financial Times (FT) AllShare index; 2) UK 12share index; 3) 12 UK stocks; 4) FT 100 index and index
futures; 5) DJIA index; 6) S&P 500 index and index futures /
Daily 
1),
2), and 3): 197291; 4):
198594; 5):
18971988; 6):
198292 
Moving
average (short
moving averages: 1, 2, and 5 days; long moving averages: 50, 100, 150, and 200, with and
without a 1%
band) 
/
Parameters are optimized for the DJIA data from 1897 to 1968. 
Not
adjusted 
The
results of optimized moving average rules indicated that differences
of mean returns between buy and sell positions were substantially positive
and statistically significant for the FTA index, all versions of the 12share
index, 4 of the 12 
18. Goodacre & Kohn Spreyer (2001) 
A random sample of 322 companies from the S&P
500 / Daily 
Prior 200 days (198896) 
CRISMA
(combination
system of Cumulative volume, RelatIve Strength,
and Moving Average) 
The S&P 500 Index / Optimized parameters 
0
to 2% per roundtrip 
The CRISMA system generated annualized profits
ranging 6.2% to 17.6% depending on transaction costs, while the annualized
return on the S&P 500 Index over the same time period was 14.2%. However, when adjusted for market movements
and risk, mean excess returns for nonzero levels of transaction costs were
significantly negative across all returngenerating models. Moreover, the results were not stable over
time, although trades on larger firms generally performed better than small
ones. 
19. Lee, Gleason,
&
Mathur (2001) 
13
Latin American spot currencies /
Daily 
199299
(various periods from data available) 
Moving
average (short
moving averages: 1 day to 9 days; long moving averages: 10 to
30 days)
and channel (2
to 50 days) 
Zero
mean profits /
Optimized trading rules 
0.1%
per roundtrip 
Outofsample
results showed that moving average rules generated significantly positive
returns for currencies of four countries: 
Table 3
continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
20. Lee, Pan, & Liu
(2001) 
9 exchange rates from Asian countries 
198894
(198995) 
The
same trading rules as in Lee, Gleason, & Mathur (2001) 
Zero
mean profits /
Optimized trading rules 
0.1%
per roundtrip 
Outofsample
tests indicated that four exchange rates from 
21. Martin (2001) 
12
currencies in developing countries /
Daily 
1/926/92 (7/926/95) 
Moving
average (short
moving averages: 1 day to 9 days; long moving averages: 10 to
30
days) 
Shortselling
strategy /
Optimized trading rules 
0.5%
per oneway transaction 
Outofsample,
moving average rules generated positive mean net returns in
10 of 12
currencies, and the returns were greater than 0.14% daily (35% per year) in
5 currencies. However, Sharpe ratios indicated that moving
average rules
did not generate superior returns on a riskadjusted basis. 
22. Skouras (2001) 
Dow
Jones Industrial Average (DJIA) /
Daily 
196286 (196286) 
Moving
average (2
to 200 days with bands of 0, 0.5, 1, 1.5, and 2%) 
Buy
& hold /
Optimized trading rules 
Various
levels from 0 to 0.1% per oneway transaction 
Outofsample returns were estimated on a daily basis. Timevarying estimated rules (by an Artificial
Technical Analyst) outperformed various fixed moving average rules employed by Brock et
al. (1992) as
well as the B&H strategy. When
considering transaction costs, however, mean returns from the optimized
trading rule were higher than the B&H mean return only after transaction costs of
less than 0.06%. 
23. Olson (2004) 
18 exchange rates / Daily 
5year insample period from 19712000 (19762000) 
Moving average (short moving averages: 1 day to 12 days; long
moving averages: 5 to 200 days) 
Buy
& hold /
Optimized trading rules 
0.1%
per roundtrip 
Outofsample results indicated that riskadjusted
trading profits for individual currencies and an equalweighted 18currency
portfolio declined over time. For the
18currency portfolio, annualized riskadjusted returns decreased from an
average of over 3% in the late 1970s and early 1980s to about zero percent in
the late 1990s. Overall, profits of
moving average rules in foreign exchange markets have declined over
time. 
Table 4 Summary
of modelbased bootstrap technical analysis studies published between 1988 and
2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1.
Brock, Lakonishok, & LeBaron (1992) 
Dow
Jones Industrial Average (DJIA) /
Daily 
18971986 
Moving
averages (1/50,
1/150, 5/150, 1/200, and 2/200 days with 0 and 1% bands) and trading range breakout (50,
150 and 200 days with 0 and 1% bands) 
Unconditional
1 and 10day returns 
Not
adjusted 
Before transaction costs, buy (sell) positions across
all trading rules consistently generated higher (lower) mean
returns than unconditional mean returns, and these results were
highly significant in most cases. For
example, a mean buy return from variable moving average rules was about 12%
per year and a mean sell return was about 7%. Moreover, the buy returns were even less volatile
than the sell returns. Simulated
series from a random walk with a drift, AR (1), GARCHM, and EGARCH
models using a bootstrap method could not explain returns and volatility of
the actual Dow series. 
2. Levich & Thomas (1993) 
5 IMM
currency
futures: mark, yen, pound, Canadian dollar, and Swiss franc /
Daily 
197690 
Filters
(0.5, 1, 2, 3, 4, and 5%) and moving average (1/5, 5/20, 1/200 days) 
Buy
& hold 
0.025%
and 0.04% per oneway transaction 
After adjustment for transaction costs and risk, every filter rule
and moving
average rule generated substantial positive mean net returns for all currencies but the
Canadian
dollar. Moreover, the
results of the bootstrap simulation indicated that, for both trading systems, the
null hypothesis that there is no information in the original time series was
rejected in 25 of 30 cases. 
3. Bessembinder & Chan
(1995) 
Asian
stock indices: /
Daily 
197591 
The
same trading rules as in Brock et al. (1992) 
Buy
& hold 
0.5,
1, and 2% per roundtrip 
Across
all markets and trading rules tested, average mean returns on buy days exceeded those
on sell days by 26.8% per year, and an average breakeven roundtrip
transaction cost for the full sample was 1.57%. In particular, technical signals generated
by the 
4. Dempsey, &
Keasey (1996) 
Financial
Times Industrial Ordinary Index (FT30) in the /
Daily 
193594 
The
same trading rules as in Brock et al. (1992) 
Unconditional
mean returns 
More
than 1% per roundtrip for large investing institutions 
Before transaction costs, buy (sell) positions across all trading
systems consistently generated higher (lower) returns than unconditional
returns. However, an extra return per
roundtrip transaction averaged across all systems appeared
to be about
0.8%, which
was relatively smaller than the roundtrip transaction costs of 1%. 
5. Kho (1996) 
4
currency futures from IMM: pound, mark, yen, and Swiss franc /
Weekly 
198091 
Moving
average (1/20,
1/30, 1/50, 2/20, 2/30, 2/50 weeks with bands of 0 and 1%) 
Unconditional
weekly mean return, Univariate GARCHM 
Not
adjusted 
Initially,
moving average rules generated substantial mean returns between 9.9% and 11.1% per
year from buy signals. These trading
returns could
not be explained by the empirical distribution of the univariate GARCHM
model as well as transaction costs or serial correlations in futures returns. However, the returns appeared to be
insignificant when timevarying risk premia, which were estimated from a
general model of the conditional CAPM, were taken into account. 
Table 4 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
6. Raj & Thurston (1996) 
Hang
Seng Futures Index of /
Daily 
198993 
The
same trading rules as in Brock et al. (1992),
without 1/150 and 2/200 moving average rules 
Unconditional
mean returns 
Not
adjusted 
Without
considering
transaction costs, average buy returns generated from both trading systems
were much higher than the unconditional oneday mean. In particular, the trading range breakout
system generated significantly higher annual returns (457% to 781%) in four out of six
rules relative to that (39%) of the B&H strategy. On the other hand, average sell returns obtained from
both systems were negative. 
7. Mills (1997) 
Financial
Times–Institute of Actuaries 30 (FT30) index in the London Stock Exchange /
Daily 
193594: 193554, 195574, 197594 
The
same trading rules as in Brock et al. (1992) 
Unconditional
mean daily return 
Not
adjusted 
For
moving average rules, each mean daily buysell return
difference
(0.081% and 0.097%) for 193554 and 195574 was much greater than corresponding unconditional
mean returns
(0.013% and 0%). For the latest
subperiod, 197594, however, the mean buysell difference was insignificantly different
from the unconditional return. Trading
range breakout rules showed similar results. None of simulated series generated by
ARARCH bootstraps earned mean buysell differences larger than the actual
difference. 
8. Bessembinder & Chan
(1998) 
Dow
Jones Industrial Average (DJIA) /
Daily 
192691: 192643, 194459, 196075, 197691 
The
same trading rules as in Brock et al. (1992) 
Buy
& hold 
Various
estimates for NYSE stocks 
The
DJIA data in this study includes dividend
payments. Over the full sample period, an average
buysell return
difference across all 26 trading
rules was 4.7%, generating a breakeven oneway transaction cost of
0.39%. However, breakeven transaction
costs have declined over time
with 0.22% for
the most recent subperiod (197691). It was compared with an estimated transaction cost
of 0.25%. 
9. Ito (1999) 
6 national equity market indices ( / Daily 
198096 for developed markets, 198896 for emerging markets 
The
same trading rules as in Brock et al. (1992) 
Buy & hold 
Nikkei index futures: 0.11% per roundtrip; other
equity indices: 0.692.21% 
After transaction costs, technical trading rules
outperformed the B&H strategy for all indices but 
10. LeBaron (1999) 
2
foreign currencies from the /
Daily and weekly 
197992 
Moving
average (1/150 days or 1/30 weeks) 
Sharpe
ratio for buying and holding on 
Commissions (0 to 0.5%) and bidask
spread (0.15%) per roundtrip 
Mean returns of the trading rule for the two currencies were statistically
significantly
different from zero. Their Sharpe
ratios (0.60 to 0.98) were also higher than those (0.3 or 0.4) for the B&H on 
Table 4 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
11. Ratner & Leal (1999) 
10
equity indices in /
Daily 
198295 
Moving
average (1/50,
1/150, 5/150, 1/200, and 2/200 days with bands of zero
and one standard deviation) 
Buy
& hold 
Various
costs from 0.15 to 2.0% per oneway transaction 
After
transaction costs, 21 out of 100 trading rules that
were applied to the 10 indexes generated statistically significant returns
(18.2% to 32.1% per year), with the profitability concentrated in four markets: Mexico, Taiwan,
Thailand, and the Philippines. When statistical
significance was ignored, however, 82 out of the 100 rules appeared
to have forecasting ability in emerging markets. 
12. Coutts & Cheung (2000) 
Hang Seng Index on the / Daily 
198597 
The
same trading rules as in Brock et al. (1992) 
Unconditional
mean returns 
Not adjusted 
Across all trading rules tested, buy (sell) signals generated significantly higher (lower) mean
returns than unconditional mean returns.
In particular, buy (sell) signals of the trading
range breakout system earned substantial average 10day cumulative return of
1.6% (5%), which was higher (lower) than that of the moving average system. 
13. Parisi & Vasquez (2000) 
/
Daily 
198798 
The
same trading rules as in Brock et al. (1992) 
Unconditional
mean returns 
1%
per oneway transaction 
Across
trading rules, mean returns on buy signals were consistently higher than
those on sell signals or unconditional mean returns. In fact, sell signals yielded negative mean
returns for most trading rules.
Although variablelength moving average rules generated significant
returns, it was unlikely that these rules were profitable if
high transaction costs were taken into account. 
14. Raj (2000) 
Yen and mark traded in Singapore International
Monetary Exchange / Intradaily 
01/199212/1993 
Filter, moving average, and channel 
Buy & hold 
0.04%
per oneway transaction 
None of technical trading rules except one rule
(2/200 moving average rule with a 1% band) generated statistically
significant returns after adjustment for transaction costs and risk. However, some trading rules appeared to
produce economically significant returns.
For instance, for the mark a 1/50 moving average rule with a 1% band
generated a riskadjusted net return of 8.8% over the twoyear period. 
15. Gunasekarage &
Power (2001) 
4 South Asian stock indices: /
Daily 
19902000 
Moving
averages (1/50,
1/100, 1/150, 1/200, 2/100, 2/150, 2/200, 5/200, and 1/50 with 1% band) 
Buy
& hold 
Not
adjusted 
For
variable
moving average rules, buy signals generated positive returns of more than
44.2% per year and sell signals generated negative returns of less than 20.8% per year. These returns, on average, were significantly different
from the B&H returns. Similar
results were obtained for fixedlength moving average rules with 10day
holding periods. 
Table 4 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
16. Day & Wang (2002) 
Dow
Jones Industrial Average (DJIA) /
Daily 
196296 
Moving
average (1/50
and 1/150 days with 0 and 1% bands) and trading range breakout (50 and 150 days with 0 and 1% bands) 
Buy
& hold 
0.05%
per oneway transaction 
Variablelength moving average rules generated daily excess returns of more than
0.027% over the B&H strategy for 196286, and all
the returns were statistically
significant. For closing levels of the
DJIA
that were estimated to reduce the effects of nonsynchronous trading, the trading rules
also outperformed the B&H, although returns were reduced
relative to previous ones and not all were statistically significant. For 198796, however,
the performance of the trading rules was inferior to the B&H strategy in most cases. 
17. Kwon & (2002) 
The
NYSE valueweighted index /
Daily 
196296: 196272, 197384, 198596 
Moving
average, combination of moving average
and momentum, and combination of moving averages for price and volume 
Unconditional
mean returns 
Not
adjusted 
Combination moving average rules of price and volume generated
the highest daily average return of 0.13% over the full sample period. Across all
subperiods but
the recent 198596
period, returns
of the trading system were statistically significantly different from unconditional mean returns. Similar results were obtained for the other two trading
systems. Simulated series from three popular
models (random
walk, GARCHM, and GARCHM with instrument variable) could not explain
returns and volatility of the technical trading systems. 
18. Neely (2002) 
4 foreign exchange rates: mark, yen, Swiss franc, and
/
Intradaily and daily 
198398 
Moving
average (1/150) 
Not
considered 
Not
adjusted 
With
daily data, the moving average rule generated positive annual mean
returns for all
series ranging from 2.4% for the Australian dollar to 8.7% for the
yen. However, when intervention periods of
central banks were removed, the trading rule returns were greatly reduced,
ranging from –2.3% to 4.5%. With intradaily data, the highest US, Swiss, and
German excess returns appeared to precede business hours and thus precede
intervention. Hence, intervention
was less likely to be a cause that generated trading rule profits. 
19. Saacke (2002) 
Dollar/mark exchange rate in the /
Daily 
197994 
Moving
average (2
to 500 days) 
Not
considered 
0.05%
per roundtrip 
Moving average rules below 170 days earned positive net returns. Bootstrapping simulations based on a random walk with drift and
a GARCH model could not account for the size of trading rule returns. Moving average rules appeared
to be highly
profitable on days when central banks intervened. However, since trading rule
returns in
periods that neither coincided with nor were preceded by interventions were also
sizable, interventions
did not seem to be the only cause of the trading rule profitability. 
Table 4 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
20. Fang & Xu (2003) 
3 Dow Jones Indexes (Industrial, Transportation, and Utilities Averages) / Daily 
18961996 
Moving average, time series models, and combination of moving average and time series models 
Buy & hold 
Various estimates 
When the market was bullish (bearish), technical
trading rules performed in general better (worse) than trading strategies
based on time series models. When a
monthly interest rate of 0.30% was assumed over the full sample period,
combination rules produced average breakeven transaction costs of about
1.01%, 1.96%, and 1.76% for the Industrial, Transportation, and Utilities
Averages, respectively, with nonsynchronous trading adjustment. These figures appeared to be substantial
improvement on those of moving average rules (0.60%, 0.84%, and 0.80%,
respectively). 
21. Sapp (2004) 
Mark and yen / Daily 
19751998 
Moving average 
Sharpe ratio for S&P500 
Bidask spread 
During the 198094 period, moving average rules
generated statistically and economically significant returns. Positive but insignificant returns after
1995 seemed to be related with a decrease in central bank intervention
activities. Transaction costs did not
affect technical trading returns except for a few shortterm trading
rules. Over the 198098 period,
annualized Sharpe ratios for a 150day trading rule and investing in the
S&P500 were 0.65 and 0.49, respectively.
However, a preliminary analysis using an international CAPM indicated
that the hypothesis that there was a timevarying risk premium in the
technical trading returns correlated with central bank interventions could
not be rejected. 
Table 5 Summary
of genetic programming technical analysis studies published between 1988 and
2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1. Neely, Weller,
&
Dittmar (1997) 
6 exchange rates: mark, yen, pound, Swiss franc, and
two cross rates (mark/yen and pound/Swiss franc) / Daily 
197577, 197880, (198195) 
100 trading rules generated by genetic programming
during each insample period 
Buy & hold / Optimized trading rules 
Insample periods: 0.1% per roundtrip;
outofsample period: 0.05% 
Outofsample, genetic trading rules generated
positive mean excess returns after transaction costs for every currency
tested. The mean excess return across
all currencies was 2.9% per year, being higher than the B&H return
(0.6%). Since betas for these trading
rule returns against various world market indices were negative, the excess
returns did not seem to be compensation for bearing systematic risk. In addition, the superior performance of
trading rules could not be explained by standard statistical models such as a
random walk, ARMA, and ARMAGARCH. 
2. Allen & Karjalainen (1999) 
S&P
500 Index / Daily 
192982 (193695) 
100 trading rules generated by genetic programming
during each insample period 
Buy
& hold 
Oneway transaction costs of 0.1, 0.25, and 0.5% 
After considering reasonable oneway transaction
costs of 0.25%, average excess returns of optimal trading rules were negative
for 9 of 10 outofsample periods.
Even after transaction costs of 0.1%, average excess returns were
negative for 6 out of the 10 periods.
In most periods, only a few trading rules indicated positive excess
returns. Overall, genetically formulated trading rules did not generate excess returns over the
B&H
strategy after transaction costs. 
3. Fyfe, Marney, &
Tarbert (1999) 
/ Daily 
198082, 198284 (198597) 
The fittest trading rule generated by genetic
programming during an insample period 
Buy & hold / Optimized trading rules 
1% per oneway transaction 
Although an optimal trading rule performed well
during the outofsample period, it appeared to have a similar structure to
the B&H strategy. When the optimal
trading rule was applied to price series bootstrapped by three popular
statistical models (a random walk, AR (1), AR (1)–ARCH (3)), only the AR (1)
model explained about 40% of the original excess trading returns. 
4. Neely & Weller (1999) 
4 cross exchange rates (mark/franc, mark/lira,
mark/guilder, mark/pound) / Daily 
197986 (198696) 
100 trading rules generated by genetic programming,
moving average (1/10, 1/50, 5/10, and 5/50 days), and filter (0.5, 1, 1.5,
and 2%) 
Buy & hold / Optimized trading rules 
Insample periods: 0.1% per roundtrip;
Outofsample period: 0.05% 
During the outofsample period, annual mean excess
returns averaged across 100 rules after transaction costs were positive for
all four currencies, ranging 0.1% for the mark/guilder to 2.8% for the
mark/pound. In contrast, moving
average rules and filter rules generated annual mean excess returns of 0.1%
and 0.2% across all currencies, respectively. There was no evidence that the excess
returns to genetic trading rules were compensation for bearing systematic
risk. 
Table 5 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
5. Wang (2000) 
S&P Index and S&P Index Futures / Daily 
198497 (198798) 
10 trading rules generated by genetic programming
during each insample period 
Buy & hold /
Optimized trading rules 
$0.50 per share + $25 per oneway transaction for
spot index; $61 per roundtrip for futures 
For S&P futures, 36 out of 120 trading rules
over the entire sample period outperformed the B&H strategy in terms of
net returns. However, the results
varied from yeartoyear. Similar
results were found when both S&P spot and futures markets were
simultaneously considered for trading.
When riskadjusted returns were assessed, 57 out of 120 rules beat the
B&H strategy. Although the
performance of trading rules was still inconsistent over sample periods, more
than 40% of the rules appeared to have some markettiming capability. 
6. Neely &
Weller
(2001) 
4 foreign exchange rates: mark, yen, pound, and
Swiss franc / Daily 
197580 (198192), 198792 (199398) 
100 trading rules generated by genetic programming
during each insample period 
Buy & hold / Optimized trading rules 
Insample periods: 0.1% per roundtrip;
outofsample period : 0.05% 
Over the period 198192, intervention information
from the Fed substantially improved the profitability of optimal trading
rules for pound and Swiss franc. For
example, the median portfolio rule increased annual excess returns from 0.5%
to 7.2% per year for the pound. In
contrast, over the 199398 period, intervention information decreased the
profitability of trading rules for all currencies but the mark. Thus, intervention activity did not seem to
be a general source of profits for technical traders. 
7. Korczak & Roger
(2002) 
24 stocks of the CAC40 Index of the Paris Stock
Exchange / Daily 
Ten 261day periods over 1/9711/99 (Ten 7day periods) 
Trading rules generated by genetic programming
during each insample period 
Two buy & hold strategies /Optimized trading rules 
0.25% per oneway transaction 
Outofsample results indicated that genetic trading
rules outperformed both B&H strategies in 9 out of 10 cases. Although newly generated trading rules
performed well over time and relative to the old rules, all rules showed good
and stable performance over the outofsample periods. No trading rule consistently performed
better than others. 
8. Ready (2002) 
Dow
Jones Industrial Average (DJIA) /
Daily 
19392000, 195762 (196386), 198186 (198700) 
50 geneticprogrammingbased trading rules and 4
moving average rules from Brock et al. (1992) 
Buy & hold, Stock/bond weighted average / Optimized trading rules 
0.13% per oneway transaction 
Moving average rules generated positive excess
returns after transaction costs for the period 196386, although they yielded
negative excess returns for the period 19872000. However, because moving average rules
performed poorly from 193962, they were less likely to be chosen by traders
at the beginning of 1963. In fact,
every genetic trading rule created over the period 195760 outperformed the moving
average rules. Similar results were
found for the period 19872000. Hence,
Ready concluded that Brock et al.’s (1992) results for the period 196386
were spurious. 
Table 5 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
9. Neely (2003) 
S&P
500 Index / Daily 
192980 (193695) 
10 trading rules generated by genetic programming
during each insample period 
Buy & hold / Optimized trading rules 
0.25% per oneway transaction 
During insample periods, genetic trading rules
generated an about 5% annual mean excess return over the B&H
strategy. During outofsample periods,
however, genetic trading rules generated negative mean excess returns over
the B&H strategy. The
riskadjusted performance based on several riskadjusted return measures was
inferior to that of the B&H strategy.
In addition, trading rules optimized by various riskadjusted criteria
also failed to outperform the B&H strategy. 
10. Neely & Weller (2003) 
4 foreign exchange rates: mark, yen, pound, and
Swiss franc / Intradaily 
2/965/96 (6/9612/96) 
25 trading rules generated by genetic programming
for each currency; 
An linear forecasting model / Optimized trading rules 
0, 0.01, 0.02 and 0.025% per oneway transaction 
There was strong evidence of predictability in
exchange rate series tested because genetically trained trading rules yielded
annual returns of over 100% with zero transaction costs in 3 of the 4
cases. However, under realistic
trading hours and transaction costs (0.025%), genetic trading rules realized
breakeven transaction costs of less than 0.02% per oneway trade in all the
exchange rates but the pound.
Moreover, genetic trading rules appeared to be inferior to the
autoregressive linear forecasting model in most cases, although their
performances were not much different. 
11. Roberts (2003) 
CBOT corn, soybean, and wheat futures / Daily 
19781998 (19801998) 
The best of ten rules optimized during each
insample period using genetic programming 
Zero profits and buy & hold 
$25 and $6.25 per contract per roundtrip for in
and outofsample periods, respectively 
Although genetically trained rules produced positive
mean net returns only for wheat futures in outofsample tests, only trading
rules that use the ratio of profit to maximum drawdown as a performance
measure generated a statistically significant mean daily net profit of $0.93
per contract. This was compared to the
B&H profit of $3.30 per contract.
For corn and soybean futures, however, genetic trading rules produced
both negative mean returns and negative ratios of profit to maximum drawdown
during the sample period. 
Table 6 Summary
of Reality Check technical analysis studies published between 1988 and 2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1. Sullivan, Timmermann, & White (1999) 
Dow
Jones Industrial Average (DJIA), S&P 500 index
futures /
Daily 
DJIA: 18971996, 18971986 (198796); S&P 500 futures: 198496 
Filter, moving average, support and resistance,
channel breakout, onbalance volume average 
Zero mean profits for mean return, a riskfree rate for the
Sharpe ratio / Optimized trading rules 
Not adjusted 
During the 189796 period, the best rule in
terms of mean return was a 5day moving average that produced an annual mean
return of
17.2% with a data snooping adjusted pvalue of zero.
The corresponding breakeven transaction cost was 0.27% per
trade. The best rule in
terms of the Sharpe
ratio generated a value of 0.82 with a Bootstrap
Reality Check
pvalue of zero, while the B&H strategy generated a Sharpe ratio of 0.034. However, during the 198796
period, the 5day moving average rule
earned a mean
return of 2.8% per year with a nominal pvalue of 0.32. Moreover, in the S&P 500 futures market, the best rule generated a
mean return of 9.4% per year with a Bootstrap Reality Check pvalue of 0.90, implying that the return
resulted from data snooping. 
2. Qi & Wu (2002) 
7 foreign exchange rates: mark, yen,
pound, lira, French
franc, Swiss
franc, and Canadian dollar / Daily 
19731998 
Filter, moving average, support and resistance, and
channel breakout 
Buy & hold, Zero mean profits / 
Adjusted 
During
the sample period, the best trading rules, which are mostly moving average
rules and channel breakout rules, produced positive mean excess returns over
the buyandhold benchmark across all currencies and had significant data
snooping adjusted pvalues for the Canadian dollar, the Italian lira, the
French franc, the British pound, and the Japanese yen. The mean excess returns were economically
substantial (7.2% to 12.2%) for all the five currencies except for the
Canadian dollar (3.6%), even after adjustment for transaction costs of 0.04%
per oneway transaction. In addition, the
excess returns could not be explained by systematic risk. Similar results were found for the Sharp
ratio criterion, and the overall results appeared robust to incorporating
transaction costs into the general trading model, changes in a vehicle
currency, and changes in the smoothing parameter in the stationary bootstrap
procedure. 
3. Sullivan, Timmermann, & White (2003) 
Dow
Jones Industrial Average (DJIA), S&P 500 index
futures /
Daily 
DJIA: 18971998, 198796; S&P 500 futures:
198496 
Technical trading systems from Sullivan et al.
(1999) and calendar frequency trading rules from Sullivan et al. (2001) 
Buy & hold / Optimized trading rules 
Not adjusted 
For the full sample period (18971998), the best of
the combined universe of trading rules, a 2dayonbalance volume strategy,
generated a mean return of 17.1% on DJIA data with a data snooping adjusted
pvalue of zero, and outperformed the B&H strategy (a mean return of
4.8%). For a recent period (198796),
the best rule, a weekofthemonth strategy, produced a mean return of 17.3%
slightly higher than the B&H return (13.6%), but the return was not
statistically significant (pvalue of 0.98).
Similar results were found for the S&P 500 futures data. Although the best rule (a mean return of
10.7%) outperformed the benchmark (mean return of 8.0%) during the 198496
period, the data snooping adjusted pvalue was 0.99. 
Table 7 Summary
of chart pattern studies published between 1988 and 2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1. Curcio, Goodhart, Guillaume,
&
Payne (1997) 
3 foreign currencies: mark, yen, and pound / Intradaily (one hour frequency) 
4/896/89, 1/946/94 
Support and resistance, highlow, minimum of the
support and low and maximum of the resistance and high, and maxmin 
Buy
& hold 
Bidask
spreads 
Across
exchange rates tested, the results of the earlier sample
period indicated that only 4 of 36
buy and sell rules yielded statistically significant positive returns after
transaction costs. Maxmin rules showed even
worse performance. For the later
period, 10 rules had positive returns but 14 rules produced
significantly negative returns. Maxmin rules all realized negative returns. 
2. Caginalp & Laurent (1998) 
All world equity closed end funds listed in Barron’s
and all S&P 500 stocks / Daily 
4/926/96, 1/926/96 
Candlestick patterns 
Average return 
Commissions ($20 for several thousand shares) and
the bidask spread (0.10.3%) 
Candlestick reversal patterns appeared to have
statistically significant shortterm predictive power for price
movements. Each of the patterns
generated substantial profits in comparison to an average gain for the same
holding period. For the S&P 500
stocks, downtoup reversal patterns produced an average return of 0.9%
during a twoday holding period (annually 309% of the initial
investment). The profit per trade
ranged from 0.56%0.76% even after adjustment for commissions and bidask
spreads on a $100,000 trade, so that the initial investment was compounded
into 202%259% annually. 
3. Chang & Osler
(1999) 
6
spot currencies: yen, mark, pound, Canadian dollar, Swiss franc, and French
franc /
Daily 
197394 
Headandshoulders, moving average (1/5, 1/20, 5/20, 5/50, and 20/50 days),
and momentum
(5, 20, and 50day lags) 
Buy
& hold, Equity
yields 
0.05%
per roundtrip 
Headandshoulders rules earned substantial returns for
the mark and yen but not for other currencies. Profits for the
mark and yen
were around 13% and 19% per year, respectively, with
being higher than
the corresponding B&H returns or 
4. Guillaume (2000) 
3 exchange rates: mark/dollar, yen/dollar, dollar/pound /
Intradaily 
4/896/89, 1/946/94 
4
trading range breakouts with a 0.1% band 
Buy
& hold 
Bidask
spreads 
For
the first sample period, several trading rules generated statistically
significant net profits, particularly, in trending markets such as
the yen/dollar market.
For the second period, however, none of the trading rules produced
significant net profits, even in trending markets. In general, supportresistance rules
performed better than MaxMin rules used in Brock et al. (1992). 
Table 7 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
5. Lo, Mamaysky, & Wang
(2000) 
Individual
NYSE/AMEX and Nasdaq stocks /
Daily 
196296 
Headandshoulders (H&S)and inverse H&S, broadening tops and
bottoms (T&B), triangle T&B, rectangle T&B, and double T&B 
Not
considered 
Not
adjusted 
Patternrecognition algorithms were used to detect 10 chart
patterns in
price series smoothed by using nonparametric kernel
regressions. The results of
goodnessoffit and KolmogorovSmirnov tests indicated that, in many cases, return
distributions conditioned on technical patterns were significantly different
from unconditional return distributions, especially, for the Nasdaq
stocks. This suggests that technical
patterns may provide some incremental information for stock investment,
even if they may not be used to generate excess trading profits. 
6. Osler (2000) 
3 foreign exchange rates: mark, yen, and pound against
U.S. dollar /
Intradaily 
1/963/98 
Support
and resistance 
Not
considered 
Not
adjusted 
“Bounce frequency” of support and resistance levels for each
currency
published by six firms was compared to that of artificial support and
resistance levels. Results
indicated that trends in intradaily exchange rates were interrupted at the
published support and resistance levels more frequently than at the artificial
ones. The results were consistent across all
three exchange rates and all six firms, although the predictive power of the
published support and resistance levels varied. Moreover, the results were
statistically significant and robust to alternative parameterizations. 
7. Leigh, Paz, & Purvis
(2002) 
The
NYSE Composite Index /
Daily 
198099 
Bull
flag charting patterns 
Buy
& hold 
Not
adjusted 
Across
all parameter combinations considered, trading rule returns in excess of the
B&H strategy were positive for all forecasting horizons (10,
20, 40, and 80 days). Moreover,
results of linear regression analyses indicated that trading rule parameters
had predictive value for both price level and future price direction. 
8. Leigh, Modani, Purvis,
& Roberts (2002) 
The
NYSE Composite Index /
Daily 
198099
(the first
500 trading days) 
Two
bull flag patterns with trading volume (a buy position is held for
100 days) 
Buy
& hold /
Optimized parameters 
Not
adjusted 
During
the outofsample period, patterns outperformed the B&H strategy. The first and the second bull flag patterns
with trading volume generated statistically significant mean returns of 14.0%
(with 55 buy signals) and 8.6% (with 132 buy signals) for 100day holding
period, respectively, while the B&H strategy profited 5.5%. 
9. Dawson & Steeley (2003) 
225 individual FTSE100 and FTSE250 stocks /
Daily 
19862001 
The same patterns as in Lo et al. (2000) 
Buy & hold 
Not
adjusted 
This study replicates Lo et al.’s (2000) procedure
on 
Table 7 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
10. Lucke (2003) 
Dollar, mark, pound, yen, and Swiss franc / Daily 
197399 
Headandshoulders 
Not
considered 
Not
adjusted 
In general, headandshoulders rules failed to
generate positive mean returns for all holding periods (1 to 15 days) except
a oneday holding period. In addition,
it appeared that trading rule profits were not correlated with central bank
intervention. 
11. Zhou & Dong
(2004) 
1451 stocks listed on the NYSE, Amex, NASDAQ / Daily 
19622000 
Headandshoulders (HS)and inverse HS (HIS), broadening tops (BT)
and bottoms
(BB), triangle
tops (TT) and bottoms (TB), rectangle tops (RT) and bottoms (RB) 
Returns for a size and momentummatched control
company 
Not
adjusted 
To reflect the uncertainty of human perception and
reasoning, fuzzy logic were incorporated into the definition of wellknown
technical patterns. For all stocks
tested, the HS, HIS, RT, and RB patterns generated significant cumulative
abnormal returns (CARs) of around 3% for 120 days. For stocks trading above $2.00, however,
the significance of CARs dramatically reduced or disappeared. The effect of small trading prices was more
severe for NASDAQ stocks. For the HS, IHS,
and RB patterns, the fuzzy logicbased algorithm appeared to detect subtly
different postpattern performances between two portfolios with different
pattern membership values. The results
for four subperiods indicated that for the RT pattern the postpattern
performances of two portfolios with different membership values were
significantly different in the first three subperiods from 1962 through
1990. This may imply that stock
markets have been efficient after the early 1990s. 
Table 8 Summary
of nonlinear technical analysis studies published between 1988 and 2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1. Gençay (1998a) 
Dow
Jones Industrial Average (DJIA) /
Daily 
196388 (Last 250 prices for each of 6 subsamples) 
Trading rules based on a feedforwad network model 
Buy
& hold / Optimized models 
$600 per roundtrip for the contract value of 1,000,000 
Trading signals as a function of past returns were generated
by a feedforward network, which
is a class of artificial neural networks. Across
subperiods, net returns of technical trading rule (7% to 35%) dominated those of
the B&H strategy (20% to 17%).
Sharpe ratio tests indicated similar results. Correct sign predictions for
the recommended positions ranged from 57% to 61% for all subperiods. 
2. Gençay (1998b) 
Dow
Jones Industrial Average (DJIA) /
Daily 
18971988
(10 most
recent prices for each of 22 subsamples) 
Trading rules based on a feedforwad network model 
An OLS model with lagged returns as regressors / Optimized models 
Not
adjusted 
In
terms of forecast
improvement measured by the mean square prediction error (MSPE), nonlinear
models (feedforward
network models) using past buysell signals from moving average rules (1/50 and 1/200) as
regressors outperformed linear specifications such as the OLS, GARCHM (1,1), and a feedforward network
regression with past returns. For 14
of 22 subperiods, the nonlinear models generated at least 10%
forecast improvement over the benchmark model. The model with a 1/50 moving average rule
provided more accurate outofsample predictions relative to one with a 1/200 rule. 
3. Gençay & Stengos
(1998) 
Dow
Jones Industrial Average (DJIA) /
Daily 
196388 (Last 1/3 of the data set for each of 6 subsamples) 
Trading rules based on a feedforwad network model 
An OLS model with lagged returns as regressors / Optimized models 
Not
adjusted 
Overall nonlinear models (feedforward network
models) outperformed linear models (OLS and GARCHM (1,1)) in terms of MSPEs
and sign predictions. The nonlinear
models with lagged returns generated an average of 2.5% forecast improvement
over the benchmark model with lagged returns.
This prediction power improved as large as 9.0% for the nonlinear
models in which past buysell signals of a moving average rule (1/200) were
used as regressors. In particular,
when the nonlinear model included a 10day volume average indicator as an
additional regressor, it produced an average of 12% forecast gain over the
beanchmark and provided much higher correct sign predictions (an average of
62%) than other models. 
4. Gençay (1999) 
5
spot exchange rates: pound, mark, yen, /
Daily 
197392
(Last 1/3 of the data set) 
Trading rules based on a feedforwad network model and
the nearest neighbor regression 
Random walk and GARCH (1,1) models / Optimized models 
Not
adjusted 
Nonlinear
models such as the nearest neighbors and the feedforward network regressions
with past buysell signals from moving average rules (1/50 and 1/200) outperformed a random walk and a GARCH (1,1) model in terms of
sign predictions and mean square prediction errors. For example, average correct sign
prediction of the nearest neighbors model was 62% for the five
currencies. Models with a 1/50 moving average rule
provided more accurate predictions over models with a 1/200 rule. 
Table 8 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
5. Fernández Rodríguez, González Martel, & Sosvilla Rivero (2000) 
The General Index of the / Daily 
196697 (10/9110/92, 7/947/95, 10/9610/97) 
A trading rule based on a feedforwad network model 
Buy & hold 
Not adjusted 
In terms of gross returns, a trading rule based on a
feedforwad network model dominated the B&H strategy for two
subperiods, while the opposite was true for most recent subperiods in which
there exists upwards trend. Correct
sign predictions for the recommended positions ranged from 5458%, indicating
better performance than a random walk forecast. 
6. Sosvilla Rivero, Andrada Félix, & Fernández Rodríguez (2002) 
Mark and yen / Daily 
198296 
A trading rule based on the nearest neighbor
regression 
Buy & hold / Optimized models 
0.05%
per roundtrip 
Trading rule generated net returns of 35% and 28%
for the mark and yen, respectively, and outperformed B&H strategies that
yielded net returns of 1.4% and 0.4%, respectively. Correct sign predictions for recommended
positions were 53% and 52% for the mark and yen, respectively, beating a
random walk directional forecast.
However, when excluding days of 
7. Fernández Rodríguez, Sosvilla Rivero, & AndradaFélix (2003) 
9 exchange rates in the European Monetary System
(EMS) / Daily 
197894, 
Trading rules based on the nearest neighbor (NN) and
the simultaneous NN regressions and moving averages (1/50, 1/150, 1/200,
5/50, and 5/200 days) 
Not considered / Optimized models 
0.05%
per roundtrip 
For most exchange rates, annual mean returns from
nonlinear trading rules based on the nearest neighbor or the simultaneous
nearest neighbor regressions were superior to those of moving average
rules. The nonlinear trading rules
also generated statistically significant annual net returns of 1.5%20.1% for
the Danish krona, French franc, Dutch guilder, and Italian lira. Similar results were found for the Sharp
ratio criterion. The nonlinear trading
strategies generated the highest Sharpe ratios in 8 out of the 9 cases. 
Table 9 Summary
of other technical analysis studies published between 1988 and 2004
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
1. Pruitt & White (1988) 
204
stocks from the CRSP at the /
Daily 
197685 
CRISMA
(combination
system of Cumulative volume, RelatIve Strength,
and Moving Average) 
Buy
& hold 
0
to 2% per roundtrip 
After
2% transaction costs and across various returngenerating models, the CRISMA
system yielded annual excess returns ranging from 6.1% to 15.1% and beat
the B&H or market index strategy.
The system also generated a much greater percentage of
profitable trading successes after transaction costs than would be expected
by chance. 
2. Schulmeister (1988) 
Mark /
Daily 
197388 
Moving
average, momentum, point & figure, combination of moving
average & momentum 
Buy
& hold 
0.04%
per oneway transaction 
All
trading rules considered produced substantial annual returns up to
16%. The combination system performed
best. The probability of an overall
loss appeared to be less than 0.005% when one of the trading rules was followed
blindly during the 197386 period. 
3. Sweeney (1988) 
14
DowJones Industrial stocks /
Daily 
195662
(197082) 
0.5%
filter rule 
Buy
& hold 
From
0.05% to 0.2% per oneway transaction 
During
the 197082
period, for
11 of 14
stocks that had earned profits before commissions in Fama and Blume’s (1966) study, a
0.5% filter rule produced
statistically significant annual mean returns after adjustment
for transaction
costs of 0.1%. For an
equally weighted portfolio of 14 stocks, the filter rule generated a mean net
return of 10.3% per year. Portfolio
returns appeared
to be robust across several subsamples but were quite sensitive to
transaction costs. 
4. 
Treasury
bond futures from CBOT /
Daily 
197887 
A
statistical pricetrend model based on ARMA(1,1) 
Buy
& hold 
0.2%
per roundtrip 
All
four trading rules generated positive average excess returns ranging from
4.4% to 6.8% per year and were superior to the B&H strategy. However, ttest results indicated that none
of the returns was significantly different from zero at the 5% level. In addition, the B&H strategy performed
better than each trading rule from 198287. 
5. Pruitt & White (1989) 
Inthemoney call options written on the 171 stocks /
Daily 
197685 
CRISMA

Not considered 
Maximum 1988 retail transaction costs 
After
transaction costs, the CRISMA system generated a mean return of
12.1% per round trip. In fact, 71.3%
of the 171 transactions were profitable after adjustment for transaction
costs. The binomial proportionality
test statistics showed that the trading profitability could not be achieved by
chance. 
6. Neftci (1991) 
DowJones
Industrials /
Monthly 
17921976 
Moving
average (150
days) 
Not
considered 
Not
adjusted 
This
study showed that moving average rules were one of the few
statistically welldefined procedures.
Trading signals of a 150day moving average rule were incorporated into a
dummy variable in an autoregression equation.
Ftest results on the variable were insignificant for 17951910 but
highly significant for 191176, indicating some predictive power of
the moving average rule. 
Table 9 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
7. Corrado & Lee (1992) 
120
stocks from the Dow Jones and S&P 500 Index /
Daily 
196389 
0.5%
ownstock filter, 0.25%
S&P 500 Index filter, 0.5%
otherstock filter 
Buy&
hold 
0.04%
per oneway transaction 
The
ownstock filter rule generated an equallyweighted mean portfolio
return of 30.8% per year during the sample period, while the B&H strategy
yielded a mean portfolio return of 11.3% per year. This difference between the returns made an
annual gross margin of 6.4% over the B&H strategy after transaction
costs. 
8. Pruitt, Tse, & White (1992) 
148
stocks and inthemoney
call options written on the 126 target stocks /
Daily 
198690 
CRISMA
(combination
system of Cumulative volume, RelatIve Strength,
and Moving Average) 
Buy&
hold 
Security: 02% per roundtrip; Option: $60 per roundtrip 
For
stocks, the CRISMA system generated annualized excess returns of between 1.0% and 5.2% after transaction costs
of 2% and outperformed the B&H or market index strategy. For options, the system generated highly
significant returns of 11.0% per option trade after transaction costs, with
63.5% of all trades being profitable.

9. Wong (1995) 
Hang Seng Index (HSI) / Daily 
19691990, 5 subperiods 
Moving average (10, 20, and 50 days) 
Buy
& hold 
Not adjusted 
In general, moving average rules performed
well. In particular, an MA10 (a 10day
moving average) bullish signal, an MA20 bullish signal, and an MA50 bearish
signal generated statistically significant excess returns over the B&H
strategy. It appeared that for buy
(sell) signals, prices declines (rises) slowly in the early preevent period
and rises (declines) sharply in the late preevent period. Prices continued to rise (declines) slowly
in the postevent period for buy (sell) signals. 
10. Cheung & Wong (1997) 
/ Daily 
198695 
Filter (0.5, 1, and 1.5%) 
Buy
& hold 
1/8
of 1% of asset value per roundtrip 
When transaction costs and risk were adjusted,
filter rules generated superior excess returns over the B&H strategy only
for the 
11. Irwin, Zulauf, Gerlow,
& Tinker
(1997) 
Futures
contracts for soybean, soybean meal, and soybean oil /
Daily and monthly 
197483
(198488) 
Channel
(40 days), ARIMA(2,0,0)
for soybean and ARIMA(1,0,1) for soybean mean and oil 
Zero
mean profits 
Not
adjusted 
During the outofsample period, the channel system
generated statistically significant mean returns ranging
5.1%26.6% for
all markets. The ARIMA models also
produced statistically
significantly
positive returns (16.5%) for soybean meal, but significantly negative
returns (13.5%) for soybeans.
For every market, the channel system beat the ARIMA models. 
Table 9 continued.
Criteria: Study 
Markets
considered /
Frequency of data 
Insample period (Outofsample period)^{} 
Technical
trading systems 
Benchmark
strategies / Optimization 
Transaction costs

Conclusion 
12. Neely (1997) 
4 foreign currencies: mark, yen, pound, and Swiss
franc / Daily 
197497 
Filter (0.5, 1, 1.5, 2, 2.5, and 3%) and moving average (1/10, 1/50, 5/10, and 5/50 days) 
Buy & hold the S&P 500 index 
0.05% per roundtrip 
Technical trading rules showed positive net returns
in 38 of the 40 cases. In general,
moving average rules performed slightly better than filter rules. Moreover, the trading profits were not
likely to be compensation for bearing risk.
For example, for the mark, every moving average rule beat the B&H
strategy of the S&P 500 Index in terms of the Sharpe ratio. The CAPM betas from the trading rules also
generally indicated negative correlation with the S&P 500 monthly returns. 
13. Goldbaum (1999) 
U.S.
TBills, a valueweighted market portfolio of all the NYSE and AMEX
securities from the CRSP, and IBM stock /
Daily 
196289 
Moving
average (1/50,
1/200, 5/50, and 5/200 days with 0 and 1% bands) 
TBill
returns 
Not
adjusted 
As a performance measure, the price error between assets was estimated using
the
nonparametric stochastic discount factor (SDF), which was either
conditioned or unconditioned on public information
(e.g. term structure). For the market portfolio returns, moving
average rules generally
had
unconditional estimates that were significantly positive or close to zero and conditional estimates
that were
negative or close to zero, implying a negative performance of the trading rules to an
informed trader. For IBM stock
returns, however, the conditional estimates on the term structure were
significantly different from zero. 
14. Marsh (2000) 
3 IMM currency futures: mark, yen, and pound
sterling / Daily 
198096, 198085 (198690), 198090 (199195) 
Markov models and moving average rules (1/5, 5/20, and 1/200 days) 
Not considered 
0.025%
and 0.04% per oneway transaction 
Before transaction costs, all moving average rules
tested yielded positive returns for both 198185 and 198690, but the rules
generated positive returns only in 3 out of 9 cases for 199195. For outofsample periods, Markov models
also generated positive returns in 2 out of 6 cases. Augmented Markov models, in which interest differentials
were included, produced substantially positive returns for all 3 currency futures
during 198690 but only for the yen during 199195. 
15. Dewachter (2001) 
4 foreign exchange rates: mark, yen, pound,
and franc /
Weekly 
197397 
Moving
average (1/30) with a 5day holding period, Markov
model and its ARMA (1,1) representation as the class of 
Not
considered 
Not
adjusted 
Across
exchange rates, the moving average rule produced a statistically significant
average return of about 6% per year and the correct sign prediction of about
55%. The extended Markov switching
model and the ARMA (1,1) representation of the Markov switching model showed
even better performance in terms of profits and sign prediction. The results of 
16. Wong, Manzur,
& Chew
(2003) 
Singapore
Straits Times Industrial Index (STII) / Daily 
19741994, Three 7year subperiods 
Moving averages and relative strength index (RSI) 
Not considered 
Not adjusted 
In general, every trading system tested produced
statistically significant returns over all three subperiods and a whole
period. Single moving average rules
generated the best results, followed by dual moving average crossover rules
and relative strength index rules. 
Table 10 The profitability of technical
trading strategies in modern studies (19882004)
Studies 
The number of studies 
Net profit range (Outofsample period) 
Comments 

Positive 
Mixed 
Negative 

A. Stock markets 



1.1%^{a} (196888) 
• For the Dow Jones Industrial Average
(DJIA) data, which was most frequently tested in the literature, results
varied considerably depending on the testing procedure adopted. In general, technical trading strategies
were profitable until the late 1980s. However,
technical trading strategies were no longer economically profitable
thereafter. • Overall, variablemoving average rules
showed a quite reliable performance for the stock market over time. • For several nonUS stock markets (e.g.,

Standard 
1 
0 
3 

Modelbased Bootstrap 
7 
2 
3 

Genetic programming 
2 
1 
3 

Reality Check 
0 
1 
1 

Chart patterns 
5 
0 
1 

Nonlinear 
3 
0 
1 

Others 
8 
1 
0 

Subtotal 
24 
5 
12 

B. Currency markets 



5%10% (197691) 
• Many studies investigated major foreign
currency futures contracts traded on the CME, i.e., the Deutsche mark,
Japanese yen, British pound, and Swiss franc.
• For major currencies, a wide variety of technical
trading strategies, such as moving average, channel, filter, and genetically
formulated trading rules, consistently generated economic profits until the
early 1990s. • Several recent studies
confirmed the result, but also reported that technical trading profits have
declined or disappeared since the early 1990s, except for the yen market. 
Standard 
7 
3 
3 

Modelbased bootstrap 
6 
0 
1 

Genetic programming 
3 
0 
1 

Reality Check 
1 
0 
0 

Chart patterns 
2 
0 
3 

Nonlinear 
3 
0 
0 

Others 
3 
1 
1 

Subtotal 
25 
4 
9 

C. Futures markets 



4%6% (197686) 
• Technical trading strategies generated
economic profits in futures markets from the late 1970s through the mid1980s. In particular, technical trading strategies
were consistently profitable in most currency futures markets, while they appeared
to be unprofitable in livestock futures markets. • Channel rules and moving average rules were the
most consistent profitable strategies.
• After the mid1980s, the profitability of technical
trading strategies for overall futures markets were not investigated
comprehensively yet. 
Standard 
5 
0 
1 

Modelbased bootstrap 
1 
0 
1 

Genetic programming 
0 
1 
0 

Others 
1 
0 
1 

Subtotal 
7 
1 
3 

Total 
58 
10 
24 


^{a} This is a breakeven oneway transaction cost of a 5/200 moving average rule, which was optimized by using the DJIA data from 1897 to 1968 (Taylor 2000).^{} 