Term
|
Definition
|
|
Term
| purpose of simple regression model |
|
Definition
| Can be used to study the relationship between two variables |
|
|
Term
|
Definition
- method of finding the linear model which minimizes the sum of the squared errors.
- Such a model provides the best explanation/prediction of the data.
|
|
|
Term
| implications of squared errors |
|
Definition
- This model seeks to avoid BIG misses
- A big u for one case leads to a REALLY big u2.
- This means regression results can be heavily influenced by outlier cases
|
|
|
Term
| assumptions of linear regression model |
|
Definition
-
Linear in the parameters
-
Random sampling
-
Sample variation in the explanatory variables
-
Zero conditional mean (what I explained on the previous slide)
-
Homoskedasticity—the variance (spread) of the error (u) is the same for any value of X
|
|
|
Term
|
Definition
-
Describes the proportion of variation in the dependent variable explained by the independent variables.
-
The typical R2 depends on what you’re trying to predict.
-
R2 tends to be lower with individual data and higher with aggregated data.
|
|
|
Term
| Effect of adding more IV's to R squared |
|
Definition
| it can never decrease when I add more IV's (it will either stay the same or increase) |
|
|
Term
|
Definition
| We assume that the unobserved error is normally distributed in the population |
|
|
