“…In contrast, the results indicate that the heteroskedasticity problem exists. It is agreed that when the heteroskedasticity exists, the OLS regression model becomes biased and fails to be the Best Linear Unbiased Estimator (BLUE) and thus the results would be unreliable and misleading (Bentes & Menezes, 2013;Ghasempour & MdYusof, 2014;Gourieroux & Monfort, 1997;Gujarati & Porter, 2009;O'Hara & Parmeter, 2013). When the problem of heteroskedasticity exists, the Generalized Least Squares (GLS) can be used as an alternative regression model (Aljandali & Tatahi, 2018;Boslaugh & Watters, 2008;Gourieroux & Monfort, 1997).…”