Porównanie metod Akaike i Hellwiga w zakresie efektywności konstrukcji modelu regresyjnego
Efficiency comparison of Akaike and Hellwig methods in constructing regression model

Summary/Abstract: This paper presents a comparison of the Akaike Information Criterion (AIC) and the Hellwig Method as methods that select explanatory variables to a model and thus enable a choice of the true model. This comparison was made in two ways. At first, the simulations were constructed in the R software (The R Project for Statistical Computing). The purpose of these simulations was to identify which of the analyzed methods more times indicates the true model. Secondly, a comparison of both methods was made on the basis of empirical data. From the set of potential explanatory variables, a set of variables was selected according to the Akaike method and the other set according to the Hellwig method. On the basis of each of the selected sets, a model was developed. Subsequently, both models were compared in terms of their adjusted coefficients of determination, standard errors of estimate and the goodness-of-fit. Results given by the simulations and results coming from the empirical models analysis indicated that the Akaike Information Criterion is a better, more efficient and more reliable tool for selecting the optimal set of explanatory variables and the true econometric model.

• Details
• Contents