Loading...

Predictability of Equity Returns over Different Horizons:Evidence from Tehran Stock Exchange

Nasiri Byrami, Leila | 2015

895 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 46939 (44)
  4. University: Sharif University of Technology
  5. Department: Economic Sciences
  6. Advisor(s): Barakchian, Mahdi
  7. Abstract:
  8. This paper aims to test an important hypothesis in financial economics: whether equity returns are predictable over various horizons? The variables that we use are dividend yield, dividend-price ratio, price-earning ratio, dividend payout ratio and stock variance. For one month horizon,we compare forecast from predictive regression and forecast from historical mean both in-sample and out-of-sample. Evidence shows that to some extent, stock variance has predictive power and predictive regression model has a better performance than the historical mean model, but fundamental variables don’t have predictive power. The multivariate model has improved the performance, but we don’t see any improving in the performance of forecast combination model. In other part of this paper we examine the predictability of long horizon returns and to make valid inference in the presence of persistency of predictive variable and overlapping observations, we use bootstrap procedure. Results show that long horizon returns are predictable from fundamental variables. Also, we show that distinction between bull/bear market has no effect on predictability of long horizon returns
  9. Keywords:
  10. Stock Return ; Tehran Stock Exchange ; Predictability ; Bull and Bear Market ; Bootstrap Statistical Inference ; Historical Mean ; Predictive Regression ; Fundamental Variables ; Long Horizon Return ; Stock Variance

 Digital Object List

 Bookmark

No TOC