A free course for students and professionals
What you will learn
To use the Pearson correlation coefficient
To build simple linear models for prediction and explanation
To build multiple linear models for prediction and explanation
To analyze the quality of different linear predictive models
Description
In this course, you will learn about how to formulate hypotheses about correlation, how to compute and interpret the Pearson correlation, the basic assumptions of linear correlation. Also, you will learn about bivariate distributions, the way we can modelate them with linear functions, and the method of adjusting the linear function to the bivariate distribution (which is the least-squares method). You will also learn how to decompose the variability of a variable that we want to predict (also called criterion variable) and different ways in which the qualities of a predictive model can be evaluated. Finally, you will learn the mathematical function of multiple linear regression, the parameters of this function, and their formulas and interpretation. We will apply all these concepts and will resolve step by step multiple regression problems. I will discuss also the multicollinearity problem and the main assumptions in multiple regression.
In all presentations, the focus is moved from mathematical aspects to basic principles behind each statistical procedure, so that each topic is very easy to understand. The rhythm of the presentation is well calibrated in order to facilitate the understanding of each new concept.
Each section is followed by several exercises in which you can consolidate your newly acquired knowledge. In addition, you will receive for each section the tables with critical values for your statistical indicators and a document with all exercises resolved in order to check if you work correctly.