Estimating and Forecasting Growth Model by Least Absolute Shrinkage and Selection Operator (LASSO): A Cross Country Analysis

Abstract

Least Absolute Shrinkage and Selection Operator (LASSO) is a method of regression analysis that uses regularization to increase prediction accuracy for the selection of variable and model interpretability. The model which is having least root mean square error is the best model. We also estimate the retention frequency for each model to show the repeatedly significant variable in each modeling for all the countries. In this study six growth models have been used for analyzing the main determinants of economic growth in case of cross countries. Time Series Data from 1980 to 2020 were used to analyze the cross country growth factors therefore, the current study looked at 43 countries with modelling these different comparative studies based on growth modelling. So, we can make six individual models and we can estimate the General Unrestricted Model with the use of econometric technique Least Absolute Shrinkage and Selection Operator. Current study found that in case of nested model or full model it is concluded that model with lag value of GDP, trade openness, population, real export, and gross fixed capital formation are the main and potential determinants to boost up the economic growth in most of the countries.