Many recent studies in technical trading rules show more potential value than earlier studies. The profitability of the technical trading rules (the filter rules and moving average rules) is investigated for the individual stocks in the first essay. Although the performances of the technical trading rules vary across securities, the 0.5% filter rule and the 10-day moving average rule generally perform well over all securities in the DJIA with one-way 0.1% transaction costs. The second essay consists of an empirical analysis on technical trading rules (the simple price moving average, momentum, and trading volume) in the stock market index. The traditional t-test is applied to examine the value of technical trading rules. The t-test is extended through using the residual bootstrap methodology under the technical trading rules utilizing random walk, GARCH-M and GARCH-M with some instrument variables. Overall, the results show that the technical trading rules add a value to capture profit opportunities over the buy-hold strategy. The generated returns from the null models does not recover the properties (mean and variance) of actual returns. The discrepancies between the simulated returns and the actual returns are large when the trading volume is considered into the technical trading rules. The limitations of test statistics, which are the independence assumption of the samples and observations, are re-solved in the Chapter VI. The application of the bootstrap on the GARCH models is demonstrated and is examined the statistical properties of the maximum likelihood estimates (MLE) on the GARCH model using (1) the parametric bootstrap, (2) non-parametric bootstrap, (3) asymptotic method. The current study uses the Monte Carlo simulation. Unlike conventional statistical methods, the bootstrap method may be relatively robust in terms of accounting for non-normality, autocorrelation, and conditional heteroskedasticity. The results show the empirical justification of bootstrap on the GARCH models.
展开▼
