KANDA DATA

Multiple linear regression analysis can determine the effect of one variable on other variables. In multiple linear regression, the number of independent variables observed is two or more than two variables. In this article, we specifically discuss multiple linear regression with two independent variables.

Multiple linear regression equations can be written in the following equation formula:

On this occasion, we will calculate the estimated regression coefficient and the intercept estimate. Another thing that needs to be known, in the regression equation, there is an error/disturbance error that accommodates the independent variables that may affect the dependent variable but are not included in this model. The assumptions must be passed in linear regression analysis using the Ordinary Least Square (OLS) method to obtain the Best Linear Unbiased Estimator (BLUE). They generally tested assumptions are multicollinearity, heteroscedasticity, autocorrelation, normality, and linearity tests.

Multiple linear regression analysis can be easily performed with the help of statistical software. However, understanding how to calculate manually is also an important thing that needs to be understood. To comprehensively understand the manual calculation of multiple linear regression analysis, I have made an audio visual divided into three parts. In each video, the results of statistical software analysis are also presented for correction of the results of manual calculations. The video also presented how the calculation and interpretation.

1. How to Calculate b_o, b₁, b₂, and R Square (R²)

An audiovisual of the calculation stages of part 1 which discusses how to calculate the value of the intercept coefficient, the estimated coefficient of the independent variable, and the coefficient of determination can be seen in the following video (video in Indonesian, please use an English translation):

2. How to Calculate SSR, SSE, df, MSR, MSE, F-value (ANOVA)

Audiovisual in the ANOVA test on the regression (video part 2), which includes the calculation of the Sum of Square value (SSR and SSE), degrees of freedom (df), Mean Square/Mean Square value (MSR and MSE) and calculated F values can be observed in the video below (video in Indonesian, please use the translation in English):

3. How to Calculate Variance, Standard Error, and t-value

Audiovisual (video part 3) on how to manually calculate variance, standard error, and t-value in multiple linear regression analysis with two independent variables can be seen in the following video (video in Indonesian, please use English translation):

Based on the video, if there are still things that need to be asked or discussed through the comments column on this website, it can be through the comments column in each of the videos. Besides being able to be used by researchers, practitioners and final year students, this manual calculation method can also be used as an additional learning supplement for lecturers and students related to the learning of Econometrics, Advanced Econometrics, Statistics Socio-Economic Statistics, Economic Mathematics, and research methodology. Finally, I hope this article will be beneficial for all of you! Want to improve your skills in statistical analysis? Check out Practical Statistics for Data Scientists: 50+ Essential Concepts Using R and Python here.