Attended: Josh, Keola

• Started Section 7 of the course: Multiple Linear Regression
• Topics learned: Multiple Linear Regression intuition, understanding p-values, null and alternative hypothesis, statistical significance, backwards elimination, forward selection, implementation in python

• COVID-19 forced today's lab hours to be online

• P-values indicate the probability that your results would appear IF you were to take the null hypothesis to be true
• Low p-value means we can reject the null hypothesis (no correlation) and accept the alternate hypothesis (correlation exists)
• We use p-values to decide which independent variables (predictors) really matter when it comes to creating a prediction model for our data
• In backwards elimination, we get rid of all of the highest p-value predictors until we are left with predictors who's p-values are small enough to be considered statistically significant.
• In forward selection, we start with the lowest p-value predictors and make our way up until we hit our threshold of statistical significance.

Contributing authors:

Created by jclaudio on 2020/09/19 21:59.