Regression and Correlation Analysis

  • Credits: 6
  • Ending: Examination
  • Range: 2P + 2C
  • Semester: winter
  • Year: 1
  • Faculty of Economic Informatics

Teachers

Included in study programs

Actuarial Science

Data Science in Economics

Teaching results

After successful completion of this class, students will be able to analyze the relationships between statistical variables through multiple regression and correlation analysis.

In particular, students will acquire the following abilities:
− Students will acquire knowledge about the concepts, principles, methods and procedures used in multiple regression and correlation analysis.
− Students will acquire knowledge about the procedures and methods to verify assumptions of a random error term in a regression, about the consequences of violating these assumptions and about solving such problems.
− Students will understand the connection between regression analysis methods and correlation analysis methods.
−
Students will acquire in particular the following skills:
− Students will be able to perform calculations for the relevant statistical procedures, both by their own calculations (especially with the use of matrix calculus), as well as with the use of professional statistical software SAS.
− Students will learn to adequately apply the procedures and methods of regression and correlation analysis and correctly interpret the results.
− They will have the ability of critical thinking in distinguishing between causal and spurious relationship and in selecting of predictors.

Indicative content

The course Regression and correlation analysis provides students with comprehensive knowledge and skills in the field of multiple regression analysis and correlation analysis, which are among the most commonly used statistical methods in the field of economics and management, both in practice and in research.

Support literature

1. Šoltés, E. (2019). Regresná a korelačná analýza s aplikáciami v softvéri SAS. Bratislava: Letra Edu.
2. Šoltés, E. (2020). Regresná a korelačná analýza s aplikáciami v softvéri SAS – zbierka príkladov. Bratislava: Letra Edu.
3. SAS Institute Inc. (2017). The REG Procedure. In SAS/STAT®14.3 User’s Guide. Cary, NC: SAS Institute Inc.
4. Wooldridge, J. M. (2013). Introductory Econometrics: A Modern Approach (5th ed.). Mason: South-Western.
5. Hebák, P., Hustopecký, J., Malá, I. (2005). Vícerozměrné statistické metody (2). Praha: Informatorium.
6. Darlington, R. B., Hayes, A. F. (2016). Regression Analysis and Linear Models: Concepts, Applications and Implementation. Guilford Publications.
7. Fox, J. (2015). Applied Regression Analysis and Generalized Linear Models. Sage Publications.
8. Belsley, D. A., Kuh, E., Welsh, R. E. (1980). Regression Diagnostics: Identifying Influential Data and Sources of Collinearity. New York: John Wiley & Sons, Inc.
9. MacKinnon, J. G. – White, H. (1985). Some Heteroskedasticity-Consistent Covariance Matrix Estimators with Improved Finite Sample Properties. Journal of econometrics, 29(3), 305-325.
Literature will be continuously updated with the latest scientific and professional titles.

Syllabus

1. Introduction to multiple regression and correlation analysis. Classical linear regression model (CLRM). Ordinary least squares estimates. 2. Overall significance of a regression and an individual contribution of explanatory variables. 3. Statistical inference for parameters of CLRM. Predictions. Confidence interval for an individual prediction and confidence interval for the expected value (mean) of the dependent variable. 4. Correlation analysis. Simple correlation (including statistical inference). 5. Multiple, partial and semi-partial correlation (including statistical inference). 6. Collinearity diagnostics. 7. Model selection methods. 8. Influence diagnostics. 9. Graphical analysis of residuals. Assumption of homoskedasticity - its verification, consequences of its violation and solution of this problem. 10. Assumption of independence and assumption of normal distribution of error term - their verification, consequences of their violation and solution of these problems. 11. Generalized linear regression model. 12. Estimation of nonlinear regression models. 13. Summary.

Requirements to complete the course

20 % assignments (2 assignments)
20 % semester project processed in SAS Enterprise Guide
60 % final exam (25% theoretical part, 35% practical part)

Student workload

Total study load (in hours): 156 hours
Distribution of study load
Lectures participation: 26 hours
Seminar participation: 26 hours
Preparation for seminars: 26 hours
Preparation for assignments: 26 hours
Elaboration of semester project: 26 hours
Preparation for final exam: 26 hours

Language whose command is required to complete the course

Slovak

Date of approval: 11.03.2024

Date of the latest change: 02.02.2022