Autocorrelation and Heteroscedasticity in Regression Analysis

Authors

  • Nand Kishor Kumar Lecturer, Trichandra Campus, Tribhuvan University, Nepal

DOI:

https://doi.org/10.3126/jbss.v5i1.72442

Keywords:

Autocorrelation, Heteroscedasticity, Regression analysis, Durbin- Waston test, Goldfeld-Quandt test, serial correlation coefficient

Abstract

In the world of econometrics and time series analysis, autocorrelation and heteroscedasticity are two statistical issues that analysts often encounter when modeling data. Both phenomena can seriously undermine the reliability of regression results if left untreated, leading to biased conclusions and faulty predictions. Autocorrelation and heteroscedasticity are important considerations in any regression analysis, especially when explaining with time series analysis. Both phenomena undermine the accuracy of the model if left unaddressed. By applying the appropriate tests and corrective measures, such as Generalized Least Squares or robust standard errors, analysts can ensure that their models remain reliable, leading to more valid inferences and better decision-making. In practice, detecting and correcting for it should be a standard part of the model validation process to avoid incorrect conclusions and improve the robustness of the results.

Downloads

Download data is not yet available.
Abstract
41
PDF
32

Downloads

Published

2023-12-31

How to Cite

Kumar, N. K. (2023). Autocorrelation and Heteroscedasticity in Regression Analysis. Journal of Business and Social Sciences, 5(1), 9–20. https://doi.org/10.3126/jbss.v5i1.72442

Issue

Section

Articles