We hope you will join us as Dr. Md. Forhad Hossain presents their research for the Department of Mathematical Sciences Colloquium Series.
“Use of Minimum Perpendicular Distance Square Method in Regression Coefficient Estimation”
January 7, 2026, 1:00 p.m., RB 118
Abstract: Apart from the traditional least square method (OLS), a new approach “Minimum perpendicular distance square method and its generalized form” with some of its relevant additional development has been taken into account in the present work to estimate the regression coefficients, with an aim to get a better fitted regression line, so that the estimated line will be the most closest line to the observed points. Mathematical form of the estimator for the parameters have been derived and a logical stand behind the relationship between the slope parameters of the lines y = ax + b and x = cy + d has been replaced. To check the advantage of our method over OLS some simulated results have been placed.