The Findings Predict stock returns based on financial ratios based on the historical cost of the least squares regression In this study, the ratios for predicting stock returns last year to predict stock returns this year has been used and the method of least squares regression (OLS) estimate the return on equity. To predict stock returns with financial ratios based on historical cost regression model (1) was used: Model (1) Rit=16.3+1.5CRit-1+10ATit-1-0 . 5 FATit-1+31.5NPMit-1-8 ROAit-1+εit
To select the type of model we shall see better Pooled model or fixed effects model, which the Chow test (F test or binding) occurs. In this test, the null hypothesis and its opposite is presented as follows:
H0: the Pooled = all intercepts are equal. H1: At least one of the intercept = fixed effects model was different. Chow test results for the model (1) are shown in Table 1