Abstract:
The purpose of this study was to examine the expected returns of Carhart model
compared to the capital asset pricing model and the implicit capital cost model
based on cash and capital returns of growth and value stocks. The statistical population
consisted of the companies listed in Tehran Stock Exchange and the time
domain is between 2007 and 2016. By choosing Cochran sampling, 126 companies
were selected as the statistical sample. The present research is an applied
research and is naturally a descriptive study. Descriptive and inferential statistics
were used to describe the data, and to analyze the data, SPSS software was used.
Also, the results showed that there is a significant difference between the mean of
total returns and returns from the capital profit of growth and value stock; while
there is no significant difference between the average cash flow of growth and
value stocks. In addition to growth stocks, the expected returns on the basis of
Carhart model are closer to real returns compared to expected returns based on the
capital asset pricing model. But about value stock, the expected returns on the
basis of Carhart model are not closer to actual returns compared to expected returns
based on the capital asset pricing model and the cost of capital, and ultimately
for growth stocks, expected returns based on Carhart model compared with
expected returns, the implicit capital cost model is closer to actual returns.
Machine summary:
The initial empirical tests of capital asset pricing model have proven their pivotal prediction based on the existence of a positive linear relationship between systematic risk (beta) and stock returns, nev- ertheless, the results of recent studies indicated that the beta coefficient as an indicator of systematic risk does not explain the difference in the average return on equity, and except the beta, there are other variables that are not within the framework of the local asset pricing model, such as firm size, book value to market value, profit/loss ratio, and financial leverage that play an effective role in ex- plaining the difference in stock returns [2].
Therefore, the above relation is not statistically significant, therefore the zero assumption for Carhart model is confirmed, that is, the average amount of estimated values is not significantly different from the actual values, and the significance level of the t-pair test for expected returns based on the capital asset pricing model for growth stocks is less than 0.
Therefore, the relationship is statistically significant, so the zero assumption for capital asset pricing model is rejected; and the significance level of t-pair test for expected returns based on the Carhart model for value stocks is less than 0.
Table 8: Results of the t-Pair Statistic of the Sixth Hypothesis (View the image of this page) Source: Researcher Findings Therefore, the average value of the implicit capital cost estimates is significantly different from the actual values, and the t-pair test for the expected returns on the basis of Carhart model for stocks is less than 0.