Home > Doc > On the Predictive Content of Technical Analysis > Concluding remarks

On the Predictive Content of Technical Analysis

Concluding remarks

This paper suggests a rationale for the application of technical analysis. In contrast to the assumptions of the standard asset market approach, information about asset price fundamentals is often only available with considerable lags. Within a more realistic informational environment, we find that the oscillator model based on moving averages is able to infer information about hidden fundamentals and can be interpreted as a cheap proxy of Bayesian learning.

The logic of technical analysis presented in this paper suggests that its forecasting success will be state dependent because it only predicts the direction of future exchange rate changes accurately when regime shifts alter the time series properties of unobservable fundamentals. However, the exchange rate is certainly not always driven by the dynamics of hidden fundamentals, implying that periods of technical forecast dominance are followed by periods of standard fundamental analysis prevalence. To examine the empirical evidence on the filter rule, a Markov regime-switching model is applied to various daily USdollar spot exchange rates. Statistically significant parameter estimates provide support for the hypothesis that the oscillator model has proven useful for forecasting purposes in low volatility periods.

Note that this is in contrast to the existing literature applying standard single regime models to evaluate the forecasting performance of filter rules (Diebold and Nason, 1990). The parameter estimates appear to be robust concerning the length of the short-run and the long-run moving average, which is in line with results of profitability studies.

Clearly, the theoretical results are based on the assumption that the market does not make use of the filter rule. But if market participants really trade on the rule, an initially given buying signal would result in an ongoing appreciation of the exchange rate. Thus it is impossible to disentangle from ex post whether a recent trend in the exchange rate, detected by the filter rule, is a result of the dynamics of hidden fundamentals or the trading activity of speculators.

Prof. Stefan Reitz

Next: Literature

Summary: Index