Exam 1

1. Metainformation

   Tag	Value
   file	Assumptions_vufsw-assumptions-1004-en_vufsw-assumptions-1004-en
   name	vufsw-assumptions-1004-en
   section	inferential statistics/regression/assumptions
   type	schoice
   solution	FALSE, TRUE, FALSE, FALSE
   Type	interpretating graph
   Program	NA
   Language	English
   Level	statistical reasoning

   Question

   Which regression assumption is notviolated according to the figures presented below?
   

   
   
   1. FALSE: The assumption of homoscedasticity because the variance of the residuals does not seem to vary for the different predicted values
   2. TRUE: The assumption of the normality of the residuals because in the P-P plot the P(Z>z) of the distribution of our data does not differ from the theoretical normal distribution
   3. FALSE: The assumption of linearity because the variance of the residuals does not seem to vary for the different predicted values
   4. FALSE: The assumption of the normality of the dependent variable because in the P-P plot the P(Z>z) of the distribution of our data does not differ from the theoretical normal distribution
   
   

   Solution

   For the assumption of homoscedasticity, we look at whether the variance of the residuals varies across the different predicted values. The variance of residuals seems to vary considerably as we move from the left to the right part of the graph. Therefore, also the assumption of homoscedasticity is violated.
   For the assumption of linearity, we would need to look at whether there is a pattern in the scatterplot and specifically whether the points are evenly spread above and below the 0-line. This is not the case, so linearity is violated.
   The assumption of normality refers to the normality of residuals and not the dependent variable. To assess the normality of residuals we use the P-P plot where our data should be close to the 45-degree line which corresponds to the theoretical normal distribution. This seems to be the case, so the assumption of normality of residuals is approximately met.

   Language
   English

   M&T MVA
   Standaardwaarde

   M&T Regression assumptions
   Default value

   
   
   1. False
   2. True
   3. False
   4. False