Multiple Regression Analysis with Python

Course details

Learn multiple regression analysis through a practical course with Python using real world data. It explores main concepts from basic to expert level which can help you achieve better grades, develop your academic career, apply your knowledge at work or make business forecasting related decisions. All of this while exploring the wisdom of best academics and practitioners in the field.

Become a Multiple Regression Analysis Expert in this Practical Course with Python

• Define stocks dependent or explained variable and calculate its mean, standard deviation, skewness and kurtosis descriptive statistics.
• Standardize rates, prices and macroeconomic independent or explanatory variables by calculating their mean and standard deviation descriptive statistics.
• Analyze multiple regression statistics output through coefficient of determination or R square, adjusted R square and regression standard error metrics.
• Examine multiple regression analysis of variance through regression, residuals and total degrees of freedom, sum of squares, mean square error, regression F statistic and regression F statistical significance.
• Review multiple regression coefficients through their value, standard error, t statistic and t statistical significance or p-value.
• Evaluate regression correct specification through individual coefficients statistical significance and correct it through backward elimination stepwise regression.
• Assess regression no linear dependency through multicollinearity test and correct it through correct specification re-evaluation.
• Appraise regression correct functional form through Ramsey-RESET test and correct it through non-linear quadratic, logarithmic or reciprocal variable transformations.
• Evaluate residuals normality through Jarque-Bera test.
• Assess residuals no autocorrelation through Breusch-Godfrey test and correct it by adding lagged dependent data as independent variables to original regression.
• Appraise residuals homoscedasticity through White, Breusch-Pagan tests and correct it through Box-Cox transformation of dependent variable.
• Test regression forecasting accuracy against random walk benchmark through mean absolute error, mean square error and root mean square error metrics.

Become a Multiple Regression Analysis Expert and Put Your Knowledge in Practice

Learning multiple regression analysis is indispensable for business analysis, financial analysis or data science applications in areas such as consumer analytics, finance, banking, health care, science, e-commerce and social media. It is also essential for academic careers in data science, applied statistics, economics, econometrics or quantitative finance. And it is necessary for any business forecasting related decision.

But as learning curve can become steep as complexity grows, this course helps by leading you through step by step real world practical examples for greater effectiveness.

Content and Overview

This practical course contains 37 lectures and 4 hours of content. It's designed for all multiple regression analysis knowledge levels and a basic understanding of Python is recommended.

At first, you'll define stocks dependent or explained variable. After that, you'll define independent or explanatory variables through their rates, prices and macroeconomic areas classification. Next, you'll calculate dependent and independent variables mean, standard deviation, skewness and kurtosis descriptive statistics. Then, you'll compute independent variables standardization.

Then, you'll analyze multiple regression statistics analysis through coefficient of determination or R square, adjusted R square and regression standard error metrics. After that, you'll analyze multiple regression analysis of variance or ANOVA through regression, residuals and total degrees of freedom, sum of squares, mean square error, regression F statistic and regression F statistical significance. Next, you'll analyze multiple regression coefficient analysis through regression coefficients value, standard error, t statistic and t statistical significance or p-value.

After that, you'll evaluate multiple regression correct specification through coefficients individual statistical significance and correct it through backward elimination stepwise regression. Then, you'll evaluate multiple regression independent variables no linear dependence through multicollinearity test and correct it through correct specification re-evaluation. Next, you'll evaluate multiple regression correct functional form through Ramsey-RESET linearity test and correct it through non-linear quadratic, logarithmic and reciprocal transformations of variables. Later, you'll evaluate multiple regression residuals normality through Jaque-Bera test. After that, you'll evaluate multiple regression residuals no autocorrelation through Breusch-Godfrey test and correct it by including lagged dependent variable data as independent variables in original regression. Then, you'll evaluate multiple regression residuals homoscedasticity through White, Breusch-Pagan tests and correct it through Box-Cox transformation or normalization of dependent variable. 

Next, you'll evaluate multiple regression forecasting accuracy by dividing data into training and testing ranges. After that, you'll use training range for fitting best model by going through steps described in previous sections. Then, you'll use best fitting model coefficient values to forecast through testing range. Finally, you'll evaluate testing range forecasted values accuracy against random walk benchmark through mean absolute error, mean square error and root mean square error metrics.