05_OrLeSq_2

Choose the correct statements

• Total sum of squares is the sum of squared deviations from the mean of y.

• Sum of squared residuals is equal to total sum of squares when the variables are demeaned.

• Sum of squared residuals always equals to zero .

• All of the above.

If the data is demeaned, then ...

• The estimator is biased, we cannot use it.

• The regression line will go through the origin.

• We need first to re-scale it in order to run regression without a constant.

• We do not need to include constant only if there is more than 1 variable in the model.

How can you predict residuals in Stata?

• reg y x, predict residuals

• First regression command, then: "predict residuals, varname"

• First regression command, then: "predict varname, res"

• predict y x, residuals

What is R squared?

• Unexplained variation over explained variation.

• Unexplained variation over total variation.

• Explained variation over unexplained variation.

• Explained variation over total variation.

Choose the correct statement:

• R-squared should be used to compare the models.

• R squared can be interpreted as the percentage of variation of y that is explained by the model.

• R squared can decrease if correlation between y and additional x is zero.

• The lower the R-squared, the better the model is.

Can R-squared adjusted be interpreted as percentage of explained variation?

• No, never

• Yes, always

• Yes, if sample size is more than 1000

• Yes, if there is just one variable on rhs