Problem:
Suppose we look at real estate sales for an older neighborhood with similar houses. Suppose the prices range from $200k to $300k. In this case, the mean and median measures will be similar. Now suppose that a house burns down and catches the house next door on fire. Before long, someone builds a large house across the two empty lots. The value of that house might be $750k. Now the mean will be much higher but the median will stay the same. Here is some data for practice:
Before data:
$200, $250, $300
Mean = median = $250
After data:
$200, $250, $300, $750
Mean = $375, Median = $275
Class,
What are the statistical definitions for:
- Outlier
- Bias
- How do these terms apply to this example?