مرکز منطقه ای اطلاع رساني علوم و فناوري

Abstract :

Regression analysis refers to methods for statistical inference about the regression function and basically, it is used in analyzing the relationship between two variables. Parametric and nonparametric regression techniques represent two different approaches to the problem of regression analysis. While parametric regression have strong assumptions and particular function form, nonparametric regression doesn’t have these requirements. Nonparametric estimator, however, are less efficient than the parametric variety when the parametric model is valid. It should be noted that even though parametric and nonparametric regression models represent different approaches to regression, this does not mean that the use of one approach precludes the use of the other. Indeed, nonparametric regression techniques can be used ases the validity of a proposed parametric model. Thus, nonparametric regression procedures may represent the final stage of data analysis or an explanatory or confirmatory step in the modelling precess. Within this context, market value/book value ratio of companies which is traded on the ISE is explained by various financial ratios. Leverage and profitability ratios are found significant statistically