Main Study—Detailed Statistical Analysis by Multiple Regression

  • R. SrinivasanEmail author
  • C. P. Lohith
Part of the India Studies in Business and Economics book series (ISBE)


Out of 150 manufacturing firms, 91 firms responded completely, which was used here for the main data analysis. After the preliminary data analysis done, the detailed statistical analysis of the collected data by multiple regression is attempted in this chapter. The first step in the detailed statistical analysis is the verification of the assumptions underlying multiple regression analysis. Linearity, constant variance (homoscedasticity) and normality are the three assumptions which will be addressed for all the individual variables. Then it proceeds to the estimation of the regression model and assessing the overall model fit.

The key takeaways for the reader from this chapter are listed below
  1. 1.

    Assumptions in multiple regression analysis.

  2. 2.

    Concept of Linearity, Homoscedasticity and Normality.

  3. 3.

    Concept of outliers and influential’s.

  4. 4.

    Concept of Multicolinearity.



Research design Multiple regression Independent variable Dependent variable Linearity Homoscedasticity Normality Scatter plot matrix Box plots Normal probability plot Skewness Kurtosis Outliers Influentials 


  1. Hair JF Jr, Black WC, Babin BJ, Anderson RE, Tatham RL (2007) Multivariate data analysis, 6th edn. Pearson Prentice Hall (Chap. 1, 2 & 4)Google Scholar
  2. Kutner MH, Nachtsheim CJ, Neter J, Li W (2005) Applied linear statistical models, 5th edn. Mc Graw HillGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2017

Authors and Affiliations

  1. 1.Department of Management StudiesIndian Institute of ScienceBengaluruIndia
  2. 2.Department of Mechanical EngineeringSiddaganga Institute of TechnologyTumakuruIndia

Personalised recommendations