Working Paper: CEPR ID: DP1480
Authors: Christopher J. Neely; Paul Weller; Robert Dittmar
Abstract: We use genetic programming techniques to identify optimal technical trading rules. We find strong evidence of economically significant out-of-sample excess returns to the rules for each of six exchange rates ($/DM, $/Yen, $/SF, $/£, DM/Yen, SF/£), over the period 1981?95. Some of the rules have a structure similar to those used by technical analysts. Betas calculated for the returns according to various benchmark portfolios provide no evidence that the returns to these rules are compensation for bearing systematic risk. ?Bootstrapping? results for the $/DM indicate that the trading rules are detecting patterns in the data that are not captured by standard statistical models.
Keywords: Technical trading rules; Genetic programming; Exchange rates
JEL Codes: F31; G15
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| genetic programming approach (C45) | optimal trading rules (G11) |
| trading rules (F14) | patterns in data (Y10) |
| returns to trading rules (G18) | not compensation for systematic risk (G19) |
| profitability of trading rules (F14) | does not stem from known statistical properties (C29) |
| application of trading rules (G18) | excess returns (D46) |
| trading rules (F14) | significant out-of-sample excess returns (G17) |
| bootstrapping results (C59) | support trading rules exploit inefficiencies (G14) |