Using Genetic Algorithms to Find Technical Trading Rules:

A Comment on Risk Adjustment

Christopher J. Neely

Original Version: September 16, 1999

Current Version: October 27, 1999

Abstract: Allen and Karjalainen (1999) used genetic programming to develop optimal ex ante
trading rules for the S&P 500 index. They found no evidence that the returns to these rules were
higher than buy-and-hold returns but some evidence that the rules had predictive ability. This
comment investigates the risk-adjusted usefulness of such rules and more fully characterizes
their predictive content. These results extend Allen and Karjalainen’s (1999) conclusion by
showing that although the rules’ relative performance improves, there is no evidence that the
rules significantly outperform the buy-and-hold strategy on a risk-adjusted basis. Therefore, the
results are consistent with market efficiency. Nevertheless, risk-adjustment techniques should be
seriously considered when evaluating trading strategies.

Senior Economist, Research Department

Federal Reserve Bank of St. Louis
St. Louis, MO 63011
(314) 444-8568 (o), (314) 444-8731 (f), [email protected]

Primary Subject Code: G0 - Financial Economics

Secondary Subject Code: G14 - Information and Market Efficiency
Keywords: technical analysis, genetic programming, trading rules, stock prices,

The views expressed are those of the author and do not necessarily reflect official positions of
the Federal Reserve Bank of St. Louis, or the Federal Reserve System. The author thanks Kent
Koch for research and programming assistance, Chuck Whiteman and Paul Weller for
comments, Franklin Allen and Risto Karjalainen for making their programs available and for
correspondence on the programs. Any errors are my own.
 Using a technique known as genetic programming (Koza, 1992), Allen and Karjalainen

(1999)—hereafter AK—searched for optimal ex ante technical trading rules on daily S&P500

data over the period 1929 through 1995. They found that the transactions cost-adjusted returns

to these rules failed to exceed the returns to a buy-and-hold strategy—despite the exclusion of

dividends from the stock return—and that the market was efficient in this sense.1 There was,

however, some evidence of predictability in returns as the rules tended to be in the market during

periods of high returns and out of the market during periods of low returns. Although AK

attributed this predictability to low order serial correlation in the stock index, they speculated

that the rules might be useful on a risk-adjusted basis despite their lower returns.

The goal of this comment is two-fold: to examine the value of genetic-programming rules

with three common methods of risk adjustment and to more fully characterize the predictability

found by the rules. This paper emphasizes that risk adjustment is not simply a secondary issue, it

is absolutely essential both for evaluating the usefulness of trading rules and for measuring the

consistency of results with market efficiency (Sharpe, 1966; Jensen, 1968; Kho, 1996; Brown,

Goetzmann, and Kumar, 1998; Ready, 1998). To evaluate risk-adjusted returns, new sets of rules

that maximize risk-adjusted measures like the Sharpe ratio (Sharpe, 1966) and the X* statistic

(Sweeney and Lee, 1990) are generated. Also, tests of market timing formally quantify the

predictability found by the rules (Cumby and Modest, 1987).

The rules fail to consistently and significantly outperform the buy-and-hold strategy by

any risk-adjusted measure. Thus, this exercise extends AK’s results to find that risk-adjusted

rule returns are consistent with market efficiency. The facts that the market indices used exclude

dividends and that some predictability may be due to spurious autocorrelation, only reinforce the

1
The return to a dynamic strategy—moving in and out of the market—will be reduced less by the exclusion of
dividends than will the return to a buy and hold strategy.

1
negative results.

METHODOLOGY

Genetic programming is a nonlinear search procedure for problems in which the solution

may be represented as a computer program or decision tree (Koza, 1992). Like its cousin, the

genetic algorithm (Holland, 1975), genetic programming uses the principles of parallel search

and natural selection to search for candidate solutions to problems of interest.2 Essentially, a

computer randomly generates a population of candidate solutions—expressible as decision

trees—to a problem of interest. The rules are required only to be well defined and to produce

output appropriate to the problem of interest—a buy/sell decision in the present case. Of course,

most of these random solutions will be quite poor, but some, purely by chance, will "fit" the in-

sample data reasonably well, generating excess returns. The computer then allows the

population to "evolve" using reproduction and mutation operators. Reproduction mixes subtrees

of the population while mutation replaces subtrees with new, randomly generated subtrees.

More fit (profitable) members of the population have a greater chance to reproduce while less fit

members have a greater chance of being replaced. In this way the genetic program searches

promising areas of the solution space by evolving a population of rules that tends to become

more adept at solving the problem in successive generations.

Genetic programming minimizes—but does not eliminate—the problem of "data

snooping" by searching for optimal ex ante rules, rather than rules known to be used by traders.

Ready (1998), for example, argues that testing rules known to be widely used by technical

traders—as done by Brock, Lakonishok and Lebaron (1992)—is a form of data snooping. This

2
practice is likely to produce spurious evidence of technical trading profits because the rules are

widely used precisely because they would have been profitable on past data.3

This paper uses programs made publicly available by AK to maintain maximum

comparability to their results.4 One difference between AK’s procedures and those used here

should be noted: Interest rates are treated differently. AK's code attributes one day’s (1/365)

interest rate to the rules during each business day—not calendar day—they are out of the market.

This practice understates the returns to the genetic programming rules by 0.5 percent or less. In

this paper, rules earn interest on calendar days—not business days—they are out of the market.5

Table 1 summarizes some of the important parameters of interest chosen by AK for their

implementation of the genetic program. AK provide more information on the program.

AK used genetic programming to construct trading rules on daily data from the S&P500

from 1929 to 1995, using ten overlapping in-sample estimation periods (1929-35, 1934-40,

1939-45… 1974-80). Each in-sample period of seven years was broken down into a training

period (five years) and a selection period (two years) to alleviate the problem of overfitting the

data. Ten independent rules were generated for each set of in-sample data. Rules with positive

excess returns over the buy-and-hold strategy in the training period were saved for out-of-sample

testing over the remainder of the data (1936-95, 1941-95…1981-95).

Each day, the trading rules generated by the genetic program observe prices and generate

a buy or sell signal indicating the position to take (the same day). The buy and sell signals are

2
Genetic algorithms require the solution to the problem to be encoded as fixed length character strings rather than as
decision trees or computer programs as in genetic programming.
3
Neely, Weller and Dittmar (1997) and Neely and Weller (1999b) have applied genetic programming to find trading
rules in the dollar foreign exchange market and the European Monetary System, respectively. Neely and Weller
(1999a) have also permitted genetic programs to use additional information—central bank intervention—as inputs to
the trading rule.
4
Programs written by Rob Dittmar produced results similar to those generated by the AK programs, suggesting that
genetic programming is robust to small change in procedures.
5
The author thanks Kent Koch for observing this and Risto Karjalainen for confirming it in private communication.

3
used along with stock prices and 30-day T-Bill interest rates to compute the continuously

compounded excess return of the rule over the return to a buy-and-hold strategy in the stock

market. This excess return over the buy-and-hold strategy at time t is given by:

 P  
(1) xsrt = (z t − 1)ln  t +1  − ln(1 + it )
  Pt  
where zt is an indicator variable taking the value 1 if the rule is in the market or 0 if the rule is in

T-Bills, Pt is the stock index and it is the interest rate on the 30-day Treasury Bill earned from

business day t to business day t+1. The cumulative excess return—also called the "fitness"—for

a trading rule from time zero to time T is the sum of the daily excess returns less a proportional

transactions cost. AK considered transactions costs of 0.1 percent, 0.25 percent and 0.5 percent.

For brevity’s sake, this comment concentrates on transactions costs of 0.25 percent.

RESULTS
Comparison with AK's Results

Table 2 shows the out-of-sample results from implementing a uniformly weighted

portfolio based on all the good rules found in-sample.6 This is similar to AK's baseline case. The

rules are assessed a 0.25 percent transactions cost for changing positions and information on day

t is used to trade the same day. As in AK (compare to Table 2, Panel A in AK), the rules

generally failed to produce positive excess returns over the buy-and-hold strategy in the sample.

With the exception of the period 1949-55, for which no good in-sample rules were found, the

out-of-sample performance was similar to that found by AK.7 While AK found only one period

in which the mean excess return over the buy-and-hold strategy was positive, the current exercise

6
Results for median portfolio rules are broadly similar to—slightly better than—those of the uniform portfolio
rules. For the sake of brevity, they will not be reported separately. The median portfolio rule goes into the market if
most of the N rules are in the market, otherwise it stays out of the market.

4
found two such periods. The rules were long in the market about 50 percent of the time and

traded 7.7 times a year, on average, though the figures varied widely with the in-sample period.

The 1974-80 period produced uninteresting rules that stayed out of the market almost all the

time.

Column 4 of Table 2 shows the mean annual return to the market when the rules are in

the market less the mean annual market return when the rules are out of the market (rb-rs).

Although there is no measure of statistical significance, positive numbers favor the proposition

that the rules have some market timing ability. While AK found that rules from 7 of 10 in-

sample periods had market timing ability by this measure, the results in this paper are slightly

more pessimistic, showing that only 5 of 9 have positive rb-rs. Because the rules' buy/sell

decisions could be closely replicated by moving average rules, AK concluded that the genetic

programming rules were taking advantage of low-order serial correlation. AK speculated that the

rules might be of use to a risk-averse speculator, but did not seriously explore that possibility.

Risk Adjustment

The criterion of judging the rules to be useful only if they generate a return that exceeds

the buy-and-hold return is neither necessary nor sufficient to conclude that the rules do not

violate the efficient markets hypothesis (EMH).8 The EMH is usually interpreted as meaning that

asset prices reflect information to the point where the potential risk-adjusted excess returns do

not exceed the transactions costs of acting (trading) on that information (Jensen, 1978). This is

potentially important because dynamic strategies, such as those found by the genetic program,

are often out of the market and therefore may bear much less risk than the buy-and-hold strategy.

7
There are two reasons why the results will not exactly replicate those found by AK: 1) Genetic programming is
inherently stochastic, generating and recombining populations probabilistically; and 2) interest rate returns were
treated differently in this analysis.

5
Although there is no universally accepted method of adjusting returns for risk, this paper will

employ three commonly used techniques: the Sharpe ratio, the X* measure, and Jensen’s α.

The Sharpe ratio—the expected excess return per unit of risk for a zero-investment

strategy (Campbell, Lo and MacKinlay, 1997)—is usually expressed in annual terms as the

annual excess return over the riskless rate to a portfolio over that excess return's annual standard

deviation. The excess return over the riskless rate to the rules at time t is given by:

 P  
(2) rt = zt ln  t +1  − ln(1 + it )
  Pt  
where zt is an indicator variable that takes the value 1 when the rule is in the market and 0

otherwise. Although the rules may have lower returns than the buy-and-hold strategy, lower

volatility may permit the returns to be leveraged up to exceed the buy-and-hold return with

similar risk. For example, if the excess return to the trading rule were only half that of the buy-

and-hold strategy, but the trading rule’s Sharpe ratio were higher, the trading rule could take

leveraged positions in the market—buying with only a 50 percent margin—to obtain equal

returns with lower risk.9 Buying with a slightly lower margin would enable the rule to obtain

higher expected returns for the same risk.

The average Sharpe ratio of the transactions cost-adjusted genetic programming rules is

about 0.02, lower than the average 0.13 Sharpe ratio—the index doesn’t include dividends—for

the buy-and-hold strategy over the ten subsamples.10 Therefore, positive returns in excess of a

buy-and-hold strategy could not be generated by leveraging up the sizes of positions held by the

8
Brown, Goetzmann, and Kumar (1998) find that risk adjustment is crucial in evaluating Dow Theory
recommendations.
9
Ready (1998) has questioned whether the strategy of leveraging returns is implementable, as the investor would
have to know—or predict—the ex post moments to compute the proper amount of leverage.
10
Jorion and Goetzmann (1999) estimate that dividends made up much of the total return to U.S. equities over the
period 1921 to 1995. The average Sharpe ratio for the buy-and-hold strategy over the 10 overlapping out-of-sample
subsamples is 0.13 while the Sharpe ratio from 1929 through 1995 is 0.06.

6
genetic programming rule.11

Of course, the rules trained on an excess return criterion may not be the best risk-adjusted

rules. To determine whether technical trading rules can produce better risk-adjusted returns than

the buy-and-hold strategy, ideally we must train a set of rules using the Sharpe ratio as the fitness

criterion. The results of this exercise are shown in Table 3. The rules trained on Sharpe ratios

failed to produce higher Sharpe ratios on average but they did show greater predictive ability by

the standard of the rb-rs and X* statistics. They also spent less time in the market (24 percent

long).

Sweeney and Lee (1990) developed another risk-adjustment strategy, the X* measure, in

the context of the foreign exchange market that may be even more appropriate for equity

markets.12 They show that, in the presence of a constant risk premium, an equilibrium daily risk-

adjusted return to a trading rule would be given by:

1 T −1   Pt +1   n  1 − c   p1 T −1
P  p T −1 
(3) X* = ∑  z t ln  + (1 − z t )ln(1 + it ) + ln  − ∑ ln t +1  + 2 ∑ ln(1 + i )
t
T t =0   Pt   2T  1 + c   T t =0  Pt  T t =0 
where zt, Pt and it are defined as before, T is the number of observations, n is the number of one-

way trades, c is the proportional transactions cost, p1 is the proportion of the time spent in the

market and p2 is the proportion of the time spent in T-Bills (p1 + p2 =1). Note that the sum of the

third and fourth terms estimates the expected return to a zero transactions-cost strategy that

randomly is in the market on a fraction p1 of the days, earning the market premium, and in T-

Bills otherwise. The risk-adjusted return—under the null of no timing ability—is the actual

11
Because dynamic strategies are at an inherent disadvantage, as the market return will, on average, exceed the
riskless return, Bessembinder and Chan (1998) pursue another strategy to compare trading rules to a market return.
They permit rules to use double leverage during periods in which they are in the market.
12
Sweeney (1988) uses the X* measure in the equity market. Ready (1998) constructs a statistic similar to Sweeney
and Lee's (1990) X*. In turn, the test statistic of X* proposed by Sweeney and Lee (1990), is virtually equivalent to
the test statistic of the coefficient β1 in the Cumby-Modest test of market timing if transactions costs are omitted
from the X* calculation.

7
return less the expected return. Positive X* statistics are interpreted as evidence of superior risk-

adjusted returns.

Most of the annualized X* statistics—net of transactions costs—in Table 2 and Table 3

are negative, indicating that the rules would not have been useful, even by this risk-adjusted

measure. Almost all the X* statistics would have been positive though, if transactions costs had

not been netted out. This supports the evidence of predictability suggested by the rb-rs statistics.

Table 4 shows the X* statistics from rules trained to maximize X* as the in-sample

fitness criteria. There are no trivial X* rules and the rules are very even handed; there are no

cases in which the rules are always in or always out of the market. The results are generally

superior to those of the rules trained on excess returns. The annualized excess return over the

buy and hold is greater than in the benchmark case and the average Sharpe ratio is about the

same as the average buy and hold Sharpe ratio over all sample periods (0.12 vs. 0.13). The mean

annualized X* statistic is also slightly positive and higher than the average X* statistics from the

rules trained with excess returns and the Sharpe ratio as the fitness criterion. However, it should

be noted that even positive X* results may be consistent with the EMH in the presence of a time-

varying risk premium.

The final risk-adjustment measure considered is Jensen’s (1968) α, the return in excess of

the riskless rate that is uncorrelated with the excess return to the market.

n 1+ c 
(4) z t [ln( Pt +1 / Pt ) − ln(1 + it )] − ln  = α + β M [ln( Pt +1 / Pt ) − ln(1 + it )] + ε t
2T  1 − c 
If the intercept in equation (4)—α—is positive and significant, then the trading rule produces

excess returns that cannot be explained by correlation with the market. To measure Jensen’s α,

returns to the market and to the trading rules were aggregated over nonoverlapping 30-day

periods and regression (4) was performed by OLS using annualized returns. Results for each set

8
of rules are shown in the 9th and 10th columns of Table 2 through Table 4. Again, the only set

of rules for which the average α is positive are those trained on X*, and these are never

significant at conventional levels.

CHARACTERIZING LOW ORDER SERIAL CORRELATION

After finding that moving average rules could closely approximate the GP rules' buy/sell

behavior, AK attributed the predictability found by their GP trading rules to "low order serial

correlation" in the returns (Campbell, Lo and MacKinlay 1997). One might speculate that a

simple time series model of returns could produce better decisions than the GP. To test this

prediction and to attempt to better characterize the nature of the predictability found by AK, a

variety of ARMA models were fit to the in-sample excess returns and the best in-sample models

and parameters were chosen by the Akaike, Schwarz and excess return criteria. The best models

were used to generate trading signals—like the genetic programs—during the out-of-sample

periods. Table 5 shows that the non-trivial ARIMA models are even less successful than the

rules constructed by genetic programming. If low-order serial correlation generates the

predictability, the genetic rules are apparently more successful at estimating it than are standard

ARIMA models.

Finally, Cumby-Modest tests of market timing ability are used to more formally

determine whether the rules have predictive content. The statistical significance of the coefficient

(β1) in the regression of excess returns on signals from the trading rule summarizes the rules’

one-day ahead timing ability:

 P  
(5) 250 ⋅ 100 ⋅ ln t +1  − ln(1 + it ) = β 0 + β1 z t + ε t .
  Pt +1  

9
 Table 6 presents strong evidence that the rules do possess predictive ability: 22 of the 25

available β1 coefficients are positive and 14 of those are significant at the 5 percent level. Of the

three fitness criteria, the X* criteria seems to have produced the rules with the most predictive

content. These results illustrate the well-known result that profitability is not necessary for a rule

to have predictive content.

CONCLUSION

This paper has investigated the results of AK (1999) to determine if ex ante optimal rules

created by genetic programming are useful on a risk-adjusted basis. Although risk-adjustment

improves the relative attractiveness of the rules, neither Sharpe ratios nor Sweeney and Lee's X*

statistic, nor Jensen’s α provide evidence that rules developed by genetic programming would

have been useful even to risk-averse speculators, contrary to AK’s reasonable speculation. Rules

trained on X* measures had the best risk-adjusted performance by all the measures,

approximately equaling the buy-and-hold return performance. Of course, risk is difficult to

measure and any risk adjustment is subject to criticism. Nevertheless, this comment argues that

trading rule results must be carefully interpreted in light of risk adjustment.

It is likely that the inclusion of dividends in the stock index, the removal of spurious

autocorrelation from the index returns, or accounting for price slippage would only strengthen

the negative results of this exercise.

10
References

Allen, F., Karjalainen, R., 1999. Using Genetic Algorithms to Find Technical Trading Rules.
Journal of Financial Economics 51, 245-271.

Bessembinder, H., Chan, K., 1998. Market Efficiency and the Returns to Technical Analysis.
Financial Management 27, 5-17.

Brock, W., Lakonishok, J., LeBaron, B., 1992. Simple Technical Trading Rules and the
Stochastic Properties of Stock Returns. Journal of Finance 47, 1731-1764.

Brown, S.J., Goetzmann, W.N., Kumar, A., 1998. The Dow Theory: William Peter Hamilton's
Track Record Reconsidered. Journal of Finance 53, 1311-1333.

Campbell, J.Y., Lo, A.W., MacKinlay, A.C., 1997. The Econometrics of Financial Markets.
Princeton University Press, Princeton, NJ.

Cumby, R.E., Modest, D.M., 1987. Testing for Market Timing Ability: A Framework for
Forecast Evaluation. Journal of Financial Economics 19, 169-189.

Holland, J., 1975. Adaptation in Natural and Artificial Systems. University of Michigan Press,
Ann Arbor, MI.

Jensen, M.C., 1968. Problems in Selection of Security Portfolios: The Performance of Mutual
Funds in the Period 1945-1964. Journal of Finance 23, 389-416.

Jensen, M.C., 1978. Some Anomalous Evidence Regarding Market Efficiency. Journal of
Financial Economics 6, 95-101.

Jorion, P., Goetzmann, W.N., 1999. Global Stock Markets in the Twentieth Century. Journal of
Finance 54, 953-980.

Kho, B.C., 1996. Time-Varying Risk Premia, Volatility, and Technical Trading Rule Profits:
Evidence from Foreign Currency Futures Markets. Journal of Financial Economics 41,
249-290.

Koza, J.R., 1992. Genetic Programming: On the Programming of Computers by Means of

Natural Selection. MIT Press, Cambridge, MA.

Neely, C.J., Weller, P.A., 1999a. Technical Analysis and Central Bank Intervention. Federal
Reserve Bank of St. Louis Working Paper 97-002B.

Neely, C.J., Weller, P.A., 1999b. Technical Trading Rules in the European Monetary System.
Journal of International Money and Finance 18, 429-458.

11
Neely, C.J., Weller, P.A., Dittmar, R.D., 1997. Is Technical Analysis in the Foreign Exchange
Market Profitable? A Genetic Programming Approach. Journal of Financial and
Quantitative Analysis 32, 405-426.

Ready, M.J., 1998. Profits from Technical Trading Rules. Unpublished manuscript. University of
Wisconsin, Madison.

Sharpe, W.F., 1966. Mutual Fund Performance. Journal of Business 39, 119-138.

Sweeney, R.J., 1988. Some New Filter Tests, Methods and Results. Journal of Financial and
Quantitative Analysis 23, 285-300.

Sweeney, R.J., Lee, E.J.Q., 1990. In: Aggarwal R., and Lee C.F. (Eds.), International
Dimensions of Securities and Currency Markets. Advances in Financial Planning and
Forecasting Series Vol. 4, Part A. JAI Press, Greenwich, CT, pp. 55-79.

12
Table 1: Genetic programming parameters of interest for AK’s implementation
Parameter AK’s Choice
Size of a generation 500
Termination criterion 50 generations or no improvement for 25 generations
Probability of selection for 2 ⋅ rank in population - 1
reproduction with rules ranked from
1 (worst) to 500 (best).
(size of population)2
arithmetic functions +, -, *, /, norm, constant between (0,2)
Boolean operators "if-then", "and", "or", "<", ">", "not", "true", "false"
functions of the data "moving average", "local maximum", "local minimum", "lag of
stock index", "current stock index"
Table 2: Uniform portfolio results from the benchmark case
# of good Annualized Annualized Annualized
In-sample rules Excess over Annualized Sharpe X* Trades In-sample Out-of-sample
period in-sample buy-and-hold rb-rs ratio statistic per year % long alpha s.e. B&H B&H
1929-35 10.00 -2.49 -13.12 -0.02 -0.46 1.85 0.17 -0.70 0.29 -8.53 6.34
1934-40 10.00 -0.83 -11.26 0.27 0.27 2.84 0.65 0.63 0.79 0.85 7.35
1939-45 7.00 -3.28 -0.64 -0.12 -1.26 4.32 0.18 -2.06 1.09 3.98 7.10
1944-50 3.00 0.64 15.41 0.29 1.12 5.45 0.81 1.82 1.06 8.01 7.52
1949-55 0.00 NA NA NA NA NA NA NA NA 15.63 6.47
1954-60 10.00 0.24 24.82 0.08 0.29 1.87 0.93 0.42 0.44 12.06 6.70
1959-65 8.00 -0.77 44.89 -0.08 -0.79 22.63 0.85 -1.15 1.28 7.31 6.28
1964-70 10.00 -1.56 24.42 -0.06 -1.29 21.54 0.70 -1.77 1.32 2.96 7.52
1969-75 10.00 -3.29 102.98 -0.18 -1.19 8.36 0.22 -1.38 0.72 -2.00 9.50
1974-80 10.00 -3.42 -5.96 NA -0.01 0.27 0.00 NA NA 4.67 9.97
mean 7.80 -1.64 20.17 0.02 -0.37 7.68 0.50 -0.52 0.87 4.49 7.47
Notes: Column 2 provides the number of rules (out of 10 trials) that had positive training period returns. Column 3 is the annualized out-of-
sample excess return, net of transactions cost, to the portfolio rule while column 4 is the mean difference between average market returns
on days that the rules were in the market and the days that they were out of the market. The portfolio mean return over the riskless rate,
net of transactions cost, divided by the standard deviation of the portfolio return is in column 5. Column 6 shows the annualized X* risk-
adjusted return measure, net of transactions cost. Columns 7 and 8 show the mean number of trades per year and the mean proportion of
time spent in the market. Jensen’s alpha and its standard error are in columns 9 and 10. The annualized buy-and-hold returns are shown in
columns 11 and 12.
13
Table 3: Results generated using the Sharpe ratio as the fitness criterion
# of good Annualized Annualized Annualized
In-sample rules Excess over Annualized Sharpe X* Trades In-sample Out-of-sample
period in-sample buy-and-hold rb-rs ratio statistic per year % long alpha s.e. B&H B&H
1929-35 10.00 -2.47 -9.25 NA -0.02 0.08 0.00 NA NA -8.53 6.34
1934-40 10.00 -2.85 113.55 0.11 -0.09 1.52 0.11 -0.08 0.34 0.85 7.35
1939-45 10.00 -2.96 6.72 -0.24 -0.62 1.66 0.05 -0.87 0.34 3.98 7.10
1944-50 10.00 -1.63 64.01 0.25 0.61 0.97 0.09 1.05 0.75 8.01 7.52
1949-55 10.00 -0.17 4.87 0.07 -0.03 2.58 0.86 -0.05 0.62 15.63 6.47
1954-60 10.00 -0.35 5.99 0.06 0.08 6.01 0.44 0.21 1.43 12.06 6.70
1959-65 10.00 -0.39 10.07 -0.07 -0.46 11.40 0.51 -0.68 1.34 7.31 6.28
1964-70 10.00 -1.32 22.20 -0.06 -0.76 9.20 0.39 -1.07 0.89 2.96 7.52
1969-75 10.00 -2.69 78.37 NA -0.01 1.01 0.00 NA NA -2.00 9.50
1974-80 10.00 -3.42 -9.97 NA 0.00 0.00 0.00 NA NA 4.67 9.97
mean 10.00 -1.82 28.66 0.02 -0.13 3.44 0.24 -0.21 0.82 4.49 7.47
Notes: see the notes to Table 2.
Table 4: Results generated using the X* measure as the fitness criterion.
# of good Annualized Annualized Annualized
In-sample rules Excess over Annualized Sharpe X* Trades In-sample Out-of-sample
period in-sample buy-and-hold rb-rs ratio statistic per year % long alpha s.e. B&H B&H
1929-35 10.00 -1.37 1.66 0.20 0.26 3.11 0.33 0.67 0.78 -8.53 6.34
1934-40 10.00 -0.77 6.62 0.29 0.70 3.19 0.53 1.29 1.17 0.85 7.35
1939-45 10.00 -2.49 -13.12 0.00 -0.93 2.34 0.37 -1.60 0.76 3.98 7.10
1944-50 10.00 -0.07 18.09 0.30 1.02 3.87 0.56 1.81 1.01 8.01 7.52
1949-55 1.00 -0.57 2.72 0.05 -0.16 4.00 0.58 -0.15 1.47 15.63 6.47
1954-60 10.00 -0.50 6.19 0.04 -0.02 5.93 0.37 0.07 1.35 12.06 6.70
1959-65 10.00 0.92 16.27 0.08 0.87 4.83 0.66 1.25 0.93 7.31 6.28
1964-70 10.00 -1.40 20.23 -0.06 -0.96 14.94 0.53 -1.34 1.08 2.96 7.52
1969-75 10.00 -2.35 -1.16 0.10 -0.25 2.13 0.22 -0.16 0.61 -2.00 9.50
1974-80 10.00 -2.23 -1.68 0.23 -0.03 2.53 0.36 0.12 0.69 4.67 9.97
mean 9.10 -1.08 5.58 0.12 0.05 4.69 0.45 0.20 0.99 4.49 7.47
Notes: see the notes to Table 2.
14
Table 5: Results from ARIMA rules
Annualized
In-sample Search AR MA Daily Excess over Annualized Sharpe X* Trades
Period Criterion Order Order Dummy buy-and-hold rb-rs ratio statistic per year % long
1929-35 AIC 5 5 2 -34.19 15.36 -3.04 -32.62 145.56 0.5750
SC 1 0 1 -3.69 NA NA 0.00 0.00 0.0000
Excess Return 1 2 1 -3.89 46.81 -0.07 -0.22 1.44 0.0030
1934-40 AIC 2 2 1 -34.40 5.21 -3.00 -32.23 134.14 0.5149
SC 2 2 1 -34.40 5.21 -3.00 -32.23 134.14 0.5149
Excess Return 1 0 1 -0.17 -2187.39 0.32 -0.17 0.04 0.9999
1939-45 AIC 5 5 3 -34.05 -10.33 -2.90 -32.27 118.79 0.5486
SC 1 0 1 -0.58 -610.46 0.25 -0.58 0.36 0.9992
Excess Return 1 0 1 -0.58 -610.46 0.25 -0.58 0.36 0.9992
1944-50 AIC 5 4 2 -25.34 13.85 -2.03 -23.78 108.27 0.6174
SC 2 0 1 -24.35 -0.26 -1.78 -23.14 92.43 0.7031
Excess Return 1 0 1 -0.20 -2187.25 0.30 -0.20 0.04 0.9999
1949-55 AIC 3 5 2 -30.43 15.35 -2.62 -29.54 131.82 0.6703
SC 2 0 2 -22.37 21.52 -1.68 -21.74 102.44 0.7669
Excess Return 1 0 1 -0.24 -1095.88 0.19 -0.24 0.10 0.9998
1954-60 AIC 4 1 3 -28.12 18.57 -2.25 -27.29 125.33 0.6850
SC 2 0 2 -24.67 21.02 -1.87 -24.00 112.11 0.7448
Excess Return 1 0 1 -0.97 -107.73 0.12 -0.96 2.06 0.9958
1959-65 AIC 2 3 2 -25.61 25.65 -2.28 -24.91 123.85 0.6256
SC 1 1 2 -23.71 22.48 -1.86 -23.27 109.16 0.7680
Excess Return 1 0 1 -7.68 -11.64 -0.46 -7.55 27.50 0.9346
1964-70 AIC 5 5 3 -33.53 13.82 -3.04 -32.16 142.47 0.5396
SC 3 2 2 -30.84 10.80 -2.71 -29.56 128.99 0.5708
Excess Return 2 2 1 -16.05 -11.26 -1.38 -14.61 47.38 0.5150
1969-75 AIC 5 5 2 -24.25 18.42 -1.96 -21.75 105.42 0.4820
SC 1 1 2 -26.49 11.81 -1.84 -25.36 109.95 0.7652
Excess Return 2 2 1 -11.67 5.59 -1.26 -7.14 29.71 0.0601
1974-80 AIC 4 3 2 -14.23 -8.02 -0.75 -11.50 38.00 0.4999
SC 2 1 1 -58.92 0.18 -5.45 -56.17 224.89 0.4951
Excess Return 4 4 1 -14.06 -8.28 -0.74 -11.34 37.07 0.5015
Notes: Column 2 shows the in-sample model selection criterion. Columns 3 and 4 show the chosen
orders of the autoregressive and moving average components. Column 5 summarizes the deterministic
component of the model: 1 indicates a simple constant, 2 indicates a weekend dummy on returns while 3
indicates that a full set of day-of-the-week dummies was used. For the other columns, see the notes to
Table 2.

15
Table 6: Cumby-Modest tests of market timing
Benchmark case Sharpe ratio rules X* rules

beta s.e. p-value| beta s.e. p-value| beta s.e. p-value|

1929-35 0.41 19.07 0.49 NA NA NA 16.11 7.53 0.02
1934-40 13.81 6.74 0.02 21.06 15.28 0.08 11.07 4.88 0.01
1939-45 -1.96 6.13 0.63 -14.41 15.84 0.82 -7.15 8.53 0.80
1944-50 23.92 6.04 0.00 83.74 19.24 0.00 23.46 6.68 0.00
1949-55 NA NA NA 12.05 9.34 0.10 3.41 4.26 0.21
1954-60 47.55 18.27 0.00 8.43 5.32 0.06 8.16 5.44 0.07
1959-65 66.34 9.54 0.00 21.10 7.70 0.00 36.51 10.83 0.00
1964-70 82.27 13.16 0.00 60.75 18.49 0.00 95.61 17.28 0.00
1969-75 69.28 28.83 0.01 NA NA NA 15.12 24.30 0.27
1974-80 NA NA NA NA NA NA 26.08 26.16 0.16
7 6 6 3 9 5

Notes: The three panels show the results of Cumby-Modest tests (see equation (5)) on the benchmark
case of excess returns, the case of rules trained on the Sharpe ratio and the X* statistic. The columns of
each subpanel show the coefficient, its standard error and its p-value. The final row displays the number
of positive betas and the number of p-values less than 0.05.

16