Question

The F-statistic tests for your whole regression

		tells us whether all the variables we are using as a group should be discarded
		tells us the sum of the t-tests of all the variables we are using
		tells us the singificance of the constant/intercept
		the concept of an F-test of the regression does not exist

Suppose you run a regression of test scores against parking lot area per pupil. Is the R2 likely to be high or low?

	A.	Low, because parking lot area is correlated with student teacher ratio, with whether the school is in a suburb or a city, and possibly with district income.
	B.	Low, Because the relationship between test scores and parking lot area per pupil is a causal relationship.
	C.	High, Because the relationship between test scores and parking lot area per pupil is a causal relationship.
	D.	High, because parking lot area is correlated with student teacher ratio, with whether the school is in a suburb or a city, and possibly with district income.

With the F-test we can check

whether two variables are a linear combination of each other

Whether the model's RMSE is positive

Whether the adj R^2 is close to 1

Whether the model's RMSE is negative

Answer 1

Question 1

The F - statistic test for whole regression is "tells us whether all the variables we are using as a group should be discarded". Option A is correct.

Question 2

It will be low, because parking lot area is correlated with student teacher ratio, with whether the school
is in a suburb or a city, and possibly with district income, it is not related with test scores.

Qustion 3

Here with F - test we check that whether two variables are a linear combination of each other.