# How much on average will cardholders who charged \$2000 in December charge in January?

Will you test goodness-of-fit or homogeneity (independence)? (.5 point)
b) Write the appropriate hypothesis. (.5 point)
c) Find the expected counts for each cell and explain why chi-square procedure are not
appropriate for this table. (1 point)
d) Create a new table by combining categories so that a chi-square procedure can be used. (1
point)

e) With this change in the table, what has happened to the degrees of freedom? (.5 point)
f) Test the hypothesis about the two groups and state the appropriate conclusion. (3.5 points)
Question 4 Seasonal spending (8 points)
Spending on credit cards decreases after the holiday spending season (as measured by amount
charged on a credit card in December). The data set in the file (Asst5_2014_Data.xlsx) contains

the monthly credit card charges of a random sample of 99 cardholders.
a) Check the conditions of a regression model for the dataset. (3.5 point)

b) Build a regression model to predict January spending from Decembers spending. (1 point)

c) How much, on average, will cardholders who charged \$2000 in December charge in
January? (.5 point)

d) Give a 95% confidence interval for the average January charges of cardholders who charged
\$2000 in December? (1.5 points)
e) From part d), give a 95% confidence interval for the average decrease in the charges of
cardholders who charged \$2000 in December? (1 point)

f) What reservations, if any, do you have about the confidence intervals you made in part d)
and e)? (.5 point)