Review DAM

Principal component analysis and Factor analysis are...

• Dimensional reduction techniques

• Predictive techniques

• Both

• None

According to Kaiser criteria, how many principal components should be retained?

• 2

• 3

• 4

• 5

Kaiser criteria states that we should choose all principal components whose eigenvalue is higher than the average variance of the original variables

• True

• False

Loadings are ....

• The variance of each principal component

• The correlation between the original variables and each principal component

• The correlation between original variables

In Factor analysis, KMO measures the sampling adequacy to the method (data should be homogeneous)

• True

• False

A factor analysis was performed with 2 factors. The output of communalities is shown below. What should we do?

• Keep the two factors

• Decrease the number of factors

• Increase the number of factors

• Do a rotation

Correspondece analysis is ...

• a method that allows to characterise the association between categorical variables

• a method used for numerical data

• a predictive method

Select the correct interpretation of a correspondence analysis result

• Large and married with kids are similar

• Large and married with kids are positively associated

• Single and married are positively associated

In cluster analysis, the hierarchical method "single" computes distance between clusters as ...

• the minimum distance

• the maximum distance

• the average distance

In K-means, we need to choose the number of clusters a priori

• True

• False