Working Paper, Federal Reserve Bank of Atlanta 97-17
In this paper we investigate the recently documented trading profits based on technical trading rules in an asset pricing framework that incorporates jump risk and time-varying risk premia. Following Brock, Lakonishok, and LeBaron (1992), we apply popular technical trading rules to the daily S&P 500 index over a long period of time. Trading profits are examined using bootstrap simulation to address distributional anomalies. We estimate a variety of asset pricing models, including the random walk, autoregressive models, a combined jump diffusion model, and a combined model of jump-diffusion and autoregressive conditional heteroskedasticity. Technical trading profits are shown to be statistically significant for the pure diffusion models and autoregressive models, yet become less significant when jump risk is incorporated into the model and virtually disappear for an asset pricing model that incorporates both jump risk and time-varying risk premia. The empirical evidence suggests that technical trading profits could be fair compensation for the risk of price discontinuity as well as time-varying risk premia of asset returns. Alternatively, technical trading profits provide a test of specification of asset pricing models; in this vein the evidence provides support for the incorporation of jump risk into asset pricing models.