On the Omission of Intercept Term in Model Estimation: A Monte Carlo Simulation Study

Ijomah Maxwell Azubuike; Nwakuya Murren Tobechukwu

doi:10.37591/rrjost.v9i1.2507

Abstract

AbstractThe issue of having an intercept term in linear regression model is an unsettled issue in the literature. It can be said that, generally, including the constant term depends on the research. But it must be known that the estimation of parameters differs according to involvement of the constant. If an intercept term exists in the model, the least squares estimate of the slope parameter will be unbiased. In this paper, we compare the results of models that include intercept and do not include intercept term on hypothetic data sets. The approaches are demonstrated via both simulation studies. Simulation-based investigation is carried out under various different scenarios using SAS 9.0 software. Results from the simulation study will be presented. We will discuss scenarios in which case they are advantageous.Keywords: Intercept, linear regression, zero-constant, slope parameter, collinearityCite this ArticleIjomah Maxwell Azubuike, Nwakuya Murren Tobechukwu. On the Omission of Intercept Term in Model Estimation: A Monte Carlo Simulation Study. Research & Reviews: Journal of Statistics. 2020; 9(1): 1-8p.