From the course: 11 Useful Tips for Regression Analysis
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Missing data
- [Instructor] Missing data are everywhere. It's an unfortunate fact of life that most real world datasets, have missing values. Individuals refuse to state their income on household surveys. Ages are not reported. Firms do not state their profit and countries report missing GDP values. But it's missing data a bad thing? Yes, it is. The general assumption is always missing data in any dataset and any regression is a likely to be a problem. The severity of it depends on what assumptions we make about the missing data. There's a few assumptions, but we often assume that data is missing for a reason. Specifically, we assume that missing data is determined by other variables. These other variables might be in our dataset or they might not be. Either way doing nothing with this kind of missing data will lead to biased estimates. So what can we do about the missing data? Quite a lot. But the main choices normally…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
Weighted regression5m 36s
-
(Locked)
Factor variables5m 40s
-
(Locked)
Polynomial variables4m 36s
-
(Locked)
Fractional variables5m 38s
-
(Locked)
Model proportions5m 39s
-
(Locked)
Centering5m 1s
-
(Locked)
Missing data5m 55s
-
(Locked)
Standardized estimates3m 22s
-
(Locked)
Graph estimates4m 9s
-
(Locked)
Contour plots3m 37s
-
(Locked)
Animate results4m 18s
-