Multiple Regression Flashcards

1
Q

Where will B1 and B2 always meet?

A

At the y-intercept (Bo).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

If we had infinite predictors, what would the last predictor be named?

A

BkXki.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the additional assumption check that must be made compared to linear regression?

A

Multicollinearity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Why is it now adjusted R-squared and not just R-squared that we report?

A

This is because we now have multiple predictors.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

How can you calculate the F-value?

A

It is the Model MS / Residual MS.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is bivariate and multivariate linearity?

A

Bivariate - two paired datasets. Testing for a relationship between them.
Multivariate - two or more variables (multiple).

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are bivariate and multivariate outliers?

A

Outliers that occur with the combination of two or more variables.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

How is Cook’s distance calculated?

A

By removing the ith data point from the model and recalculating the regression. It summarises how much all values of the model change when the ith term is removed.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is multicollinearity?

A

Where two or more predictors in a multiple regression model are highly related with each other.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is perfect multicollinearity?

A

When the correlation between two predictors is 1.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What should VIF be?

A

It should be less than 10 for there to be no multicollinearity.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How do you interpret the standardised results of a continuous predictor?

A

The change in the outcome depends on … SD increase in predictor.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How do you interpret the results of a dichotomous predictor (dummy coded predictor)?

A

Use means and p-values.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is a 95% confidence interval?

A

A 95% CI is a range of values which contain the true value of the population with 95%.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How should you report Adjusted R-squared?

A

In % terms.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Why is the F-value sometimes not very useful?

A

Because if the predictor is significantly predicting the outcome variable, then the F-ratio will generally be larger than 1 too.
Most of the time; F-value does not tell us more than what we would have learnt by simply looking at the p-values for each predictor.