Estimating the Parameters of a Simple Linear Regression Model Without Using Differential Calculus
To estimate the parameters of a simple linear regression model, students who already know calculus can minimize the total squared deviations by setting its first-order partial derivatives to zero and solving simultaneously. For students who do not know calculus, most teachers/textbooks simply state the formulas without justifying them. Students accept the formulas on faith; and for given data, they evaluate the estimates using a calculator or a statistical software. In this paper, we justify the formulas without invoking calculus. We hope the users of statistics will benefit from our proposed justifications.
Copyright (c) 2022 Jyotirmoy Sarkar, Mamunur Rashid
This work is licensed under a Creative Commons Attribution 4.0 International License.