please answer all

Question 1

Suppose we have variables:

Spending, measured in dollars

Rewards (dummy variable with values 0 or 1, where 0 represents "did not sign up for rewards" and 1 represents "did sign up for rewards")

Suppose we would like to predict the average sales per customer (in dollars) based on if a customer is part of the customer rewards program. Which method would be more appropriate to use?

	linear regression
	logistic regression

Question 2

Suppose we have the following data about customers are a large store:

Group: A or B

Churn: Yes or No

We compute a two proportion Z-test, and find a p-value of 0.82

We can conclude that we are 95% confident that

	the average number of customers who churn is about the same in both groups.
	the proportion of customers who churn is different between Group A and Group B.
	the average number of customers who churn is different when comparing the two groups.
	the probability that a customer churns is about the same in both groups.

Question 3

Suppose we have a data set. The data comes from a sample of 17,500 individuals, and the data set consists of 2 variables:

Location: There are 6 different possible values for the variable Location.

Group: There are 5 different possible values for the variable Group.

Suppose we compute Pearson's Chi-Squared test to determine if there is a relationship between Group and Location. When we compute statistics for the Chi-Squared test, we need to use a specific value for "degrees of freedom" in order to compute the critical number. In this situation, the degrees of freedom = _____ .

Question 4

Suppose we wish to calculate the probability that a customer will churn. We have the following variables:

Churn: (values are 1 and 0: 1 for "yes" and 0 for "no").

X: quantitative independent variable

We use RStudio to compute the logistic regression model below:

Model <- glm(Churn ~ X, family="binomial")

summary(Model)

We obtain the coefficients below.

Variable	Estimate of coefficient	p-value
(intercept)	10	Less than 0.0001
X	-0.5	Less than 0.0001

When X = 18, the probability that a customer will churn equals ____ %.

(Rounding: use 3 or more decimal places for any intermediate calculation that includes decimal places.)

Question 5

You will need RStudio to complete this question.

Suppose you would like to know if the likelihood of a customer churning is related to their customer segment (A, B, or C).

Summary data from a sample is given below.

	Churn	Did not churn
Segment A	20	500
Segment B	30	1000
Segment C	10	200

We can compute Pearson's Chi-Squared test to determine that there isn't a statistically significant relationship between the likelihood that a customer churns and customer segment. The p-value that we would use to make this conclusion is ______. (Use at least 4 decimal places in your answer)