Linear Regression: Meaning of Confidence Intervals for Slope and Intercept

Organized by textbook: https://learncheme.com/ When you perform a linear regression, what you're implicitly assuming is that there is some reason that the data are linearly related. The goal of a regression analysis is to identify the true slope and the true intercept (which you never get to know in real life). If you performed an experiment 100 times and collected 100 different data sets, and then calculated the 95% confidence intervals for the slope and the intercept, on average 95 of your confidence intervals would have contained the true values of the slope and the intercept. However, you never get to know which of your 5 confidence intervals did not contain the true values. This screencast presents some simulations to show how that might play out in real life. Made by faculty at the University of Colorado Boulder, Department of Mechanical Engineering and produced by the Department of Chemical & Biological Engineering.
Check out our Engineering Computing playlists: https://www.youtube.com/user/LearnChe...