Tutoria 3

The principle that “a simpler model is be preferred over a more complex model” is called

• Simpson's paradox

• Central limit theorem

• Ockham’s Razor

• None of the above

A regression model in which more than one independent variables are used to predict the dependent variable is called as _________.

• Logistics regression

• Simple linear regression

• Multiple linear regression

• Clustering

In linear regression, residuals refer to:

• the difference between the predicted Y and the mean of observed data

• the difference between the predicted Y and actual Y values

• the difference between the predicted Y and the mean of predicted Y values

• the mean absolute error

We want the residuals to be:

• Centralised in the mean

• As close to 0 as possible, with no apparent pattern

• As close to 0 as possible

• Skewed to the left

Which of the following is the loss function for a linear regression model:

• Mean absolute error

• Root mean square error

• Root mean squared error

• The sum of squares of residuals

Select the metrics used to evaluate regression models

• Area under the ROC curve

• Root mean error

• Root mean square error

• Precision

Select the statements that are true

• An influential point changes the estimates proportionally when omitted.

• Influential points always have high leverage

• Leverage depends on xi but not on yi. It depends on how far away xi is from the mean of the xi’s

• Leverage depends on yi but not on xi. It depends on how far away yi is from the mean of the yi’s