The Comparison of Some Methods in Analysis of Linear Regression Using <i>R</i> Software

Abstract This article contains the OLS method, WLS method and bootstrap methods to estimate coefficients of linear regression their standard deviation. If holds random errors with constant variance if those are independent normally distributed we can use least squares which is accurate for drawing inferences these assumptions. heteroscedastic, meaning that depends from explanatory variable, or have different weights, can’t because this cannot be safe results. know weights each error, weight method. In also described evaluate parameters. The improved quantile estimation. We simulated non variances in a using R program comparison Using software found confidence interval, estimated coefficients, plots results any case.