Quality of Linear Regression Basing on Assumption About Distribution of Standardized Residuals: Analysis of Skewness and Kurtosis

Student: Mishugina Maria

Educational Programme: Applied Methods of Social Analysis of Markets (Master)

Year of Graduation: 2019

This study investigates distributions of standardized residuals in OLS – regressions. The issue is that researchers usually do not know how far they are allowed to violate normality of residuals without loosing in quality of their models. Therefore, they either ignore the problem emerged or switch to more complicated models, loosing one of the most useful tool to predict independent variable. Key objective of the study is to propose boarders and practical recommendations of main steps in solving the problem. As a result, paper illustrates that models with different level of prediction have different stability towards kurtosis and skewness of residual distribution. Basing on this, study suggests recommendations on how to work with different types of abnormal distributions of residuals in linear regression.

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses