Questions 8-9: You have fitted the regression model to predict house selling price. Two numeric variables, AboveSpace and Age, were included, as well as two binary variables. PoorCondition = 1 if the house is in very poor condition and GoodCondition = 1 if the house is in very good condition. Otherwise, both variables equal 0. Houses in average condition would be score 0 for both variables.

Question 8

The coefficient for PoorCondition is -33351.47. This means that

a

A house in very poor condition would be worth $33,351 less than a house in very good condition.

b

A house in very poor condition would be worth $33,351 less than the average house.

c

The average house in poor condition would be worth $33,351 less than the average house of the same size and age.

d

All of the above.

Question 9

The average price of a 20 year old house with 1,500 square feet of space that is in good condition is $178,824. Approximately, within what range can we expect a particular most homes like this to sell for?

a

$178,824.

b

$122,063 and $235,585.

c

$148,968 and $208,681

d

The model estimates only the average price.

Question 10

Classification models are

a

An example of Supervised Learning.

b

The same as Clustering models.

c

Best solved using Multiple Regression.

d

None of the above.

Question 11

Applying multiple regression to classification presents challenges because

a

The outcome is binary.

b

Only gives estimates of the probability of being in a group (class).

c

Can give probability outcomes that are not between 0 and 1.

d

All of the above.