PREDICTING STOCK RETURNS USING FUNDAMENTAL INFORMATION AND MULTIVARIATE STATISTICAL MODELLING: AN EMPIRICAL STUDY ON ISTANBUL STOCK EXCHANGE

Year 2000, Volume: 18 Issue: 2, 433 - 450, 31.12.2000

Ramazan Aktaş , Mehmet Baha Karan

Abstract

This
paper performs a financial analysis that combines a set of fundamental
information into a summarv measure which predicts the return of
stocks bv usin2 logit analysis, The findings suggest that the predictive
power of financial ratios is verv high and more important than the fundamental
information. but the variables (ratios) of logit models are not stable
from one period to another. Also it is found that there is a statistically significant correlation benveen the observed and predicted
ranking. We conclude that developing a more general model for prediction might
solve the problem about unstable variables, but the general
model has verv limited ability of ranking the stocks according to their
perfonnance.

Keywords

Fnancial statement analysis, logit model, prediction of stock perfonnance

References

• Ağaoğlu, E. A. (1989) Türkiye' de Banka İşletmelerinin Ekonomik Analizi ve Gelişme Eğilimleri. Unpublished Ph.D. Thesis. Ankara.