Things to keep in mind

•  Today’s practice questions reflect the style of SOME questions in the exam.

•  It does not imply that different style of questions will not be asked.

•  Difficulty does not necessarily reflect the difficulty of questions in the exam.

•  This practice is just intended to further help you understand the methods.

• You must still study and understand the content of the whole unit. Not just practice the questions in this lecture.

Example 1

A hospital has determined that the probability that emergency room (ER) patients have heart disease is given as

Pr(yi  = 1|xi ) = g(4 + 0.02Agei  − 0.04MaxHRi ).

Where MaxHR denotes the maximum heart rate of the patient measured in a span of two minutes.  When

the probability that a patient has heart disease is greater that 80%, the patient is immediately seen by a doctor. A 38 year old patient has been recently admitted to the ER. For what values of MaxHR would this patient need to be seen immediately by a doctor?

Example 2

The data set in the plot below consists of the two predictors Walks and HmRun for 10 Baseball players.

We want to fit a regression tree to this data in order to predict salary. We want to consider a tree with two leaves (or terminal nodes). The annual salary for these players is

1. According to the overall SSE measure, which of the following is the best tree partition for this data?

Partition  1  =    {j  = Walks, s  = 50};  Partition  2  =    {j  = Walks, s  = 35};  Partition  3  =    {j  =

HmRun, s = 10}; Partition 4 =   {j = HmRun, s = 20}.

2. Using the optimal partition, what is the predicted salary for the player John  Doe who has produced

15 home runs and a total number of 55 walks?