From the course: Practical Python for Time Series Analysis

Unlock this course with a free trial

Join today to access over 25,500 courses taught by industry experts.

Robust regression for assumption violations

Robust regression for assumption violations - Python Tutorial

From the course: Practical Python for Time Series Analysis

Start my 1-month free trial Buy for my team

Robust regression for assumption violations

“

- [Instructor] Remember that we must validate these assumptions because from the numbers we get on the linear regression, on the summary, we have many P values and numbers that may be given false conclusions because some of the assumptions are unmet. We continue with the normality of residuals. In these plots, we have the residuals that we see in the table, which is the difference between real data and the predictions. And on average, most of them should be around zero with a bell curve shape that will hold the normality of these numbers. And here is what we get, which is kind of true. There is few data points. That's why we have maybe not so much, but again, we must validate the test with the hypothesis testing. There is another plot that says theoretical quantiles on the x axis sample quantiles, which is the real data on the Y. And if there is a line following the pattern of the points, we can say that there is a…

Contents