Abstract

In earlier work we examined a GARCH-M model of Turkey's Securities Markets, using it to conduct two tests of the Efficient Market Hypothesis in these markets. We found a surprising result - that we could form portfolio return estimates from a modified CAPM that led to the conclusion that the highest expected return portfolio of Turkish and world securities varied over time In subsequent research we demonstrated that a specific trading rule based on these results led to very high dollar profits in monthly transactions over a 15-year period. We now examine possible explanations for this result. We find that our in-sample result was due primarily to the model's identification of the impact of the real risk in Turkey's security markets on our portfolio's expected returns. We consider the relevance of our results to testing of the Efficient Market Hypothesis, usually identified with a buy-and-hold strategy rather than the active trading strategy of our analysis. We argue that our results are nonetheless consistent with the Efficient Market Hypothesis, in the presence of market segmentation. Finally, using a bootstrap test, we examined whether trading profits were robust to changes in the historical evolution of the data under the null hypothesis that the model was a true description of the behavior of the portfolio and the distribution function of residuals was the population distribution of forecasting errors. We found that the performance of trading profits was not robust under these conditions. We conclude that there is little doubt that the volatility of markets in Turkey is an important part of the evolution of Turkish common stock returns series. However we also see that although using this information can lead to higher expected returns to market participants, expecting to generate above average returns based on price information from our model is a risky enterprise. Unlike the apparent behavior of the long run market risk premium in the United States, it appears that the unique risks of Turkey's securities markets do not decline in the long run.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call