Use the generalized maximum Tsallis entropy estimator to fit the regression model to the data that have collinearityproblem

Sometimes a regression dataset may present us with limitations. In otherwords, they do not have a series of conditions that must have be met. In these situations,the estimates do not have enough accuracy and their variance is very large and the OLSestimator performs poorly. On the other hand. because of the non-existence of the underlyingassumptions of regression, The method of OLS cannot be used. The generalizedmaximum entropy method does not require establishing any underlying hypothesis. Tsallisentropy is nonadditive, in this paper Generalized Maximum Tsallis Entropy estimatoris introduced and through a numerical example, an application of the new estimator isillustrated. Using GMET۲, a regression model is estimated and three estimators (GME,OLS, GMET۲) are compared. Keywords: Tsallis Entropy, Generalized Maximum TsallisEntropy, Multicollinearity, Regression.