The Estimation of Cobb-Douglas Production Function Parameter through A Robust Partial Least Square

Abstract

The paper discusses the development of the agricultural production model for the agricultural sector based on the model suggested by the Libyan Government underlining the country policy in promoting this sector. The development of this agricultural production model is through the method of Cobb Douglas production function. As a step of precaution, the robust partial least squares is incorporated in the Cobb Douglas production model to avoid the effect of multicollinearity and outliers in the data. The new model is known as Robust Partial Least Squares of Cobb Douglas Production Function (RPLS-CDF). This study shows that the RPLSCDF overcomes the problem in Cobb Douglas production model when multicollinearity and the outliers exist in the data as compared to the conventional method namely the log transformation. It is found that the pernicious effect of the outliers have been reduced considerably by the proposed robust estimators.