Statistics for BA II

Which of the following methods is a unsupervised learning method?

• K-Means

• random forest

Adjusted Rand Index measures

• Goddness of the clustering solution

• Agreement between two different clustering solutions

In logistic Regression we estimate the parameters by

• ordinary leaste squares

• maximum likelihood

• None of the two previous methods

The link function refers to

• the way we calculate distance between clusters

• the function of the mean we wish to model

the log link function is used in

• logistic regression

• linear regression

• none of the previous

Solution of K-means depends on

• The initial values

• the order of data processing

• Both

Mahalanobis distance is suitable for

• Continuous data

• Mixed data

If we have 1000 observations and we apply 5-fold cross validation how many times we have to fit the model?

• 5

• 200

Regression tree is used

• for creating a classification rule

• for clustering

• None of the previous

Matching similarity coefficient relates to

• Euclidean Distance

• Manhattan distance

• Both