No. 25 (enero junio 2017)

Browse

Now showing 1 - 1 of 1

Bases for alternative nonparametric mincer function
(Universidad de Cuenca, 2017-01) Sacoto Molina, Matias Nicolas; Brondino, Alejandro
This work undertakes a nonparametric regression in order to assess the viability of this technique in modeling a simplified Mincer Function of earnings applied to the NBA players’ wages. The main advantages of using this technique is that it does not rely on assumptions and the statistical inference is not sensitive to distributions disturbances due to violations of the assumptions. The results of the nonparametric estimation are compared to a classical OLS regression. We found evidence that the OLS estimator did not fulfilled the assumptions that this method requires, therefore, the statistical inference form this estimation could lead to wrong conclusions (due to lack of efficiency), unless some correction that solves the violation to the assumptions is applied to the model. On the other hand, the confidence intervals obtained from the nonparametric regression are more accurate and less sensitive to variability and magnitude of the variables. Consequently, the nonparametric estimation would be an alternative to model the behaviour of the wages avoiding strong assumptions that could lead to wrong statistical inference conclusions.