# Question

Recall from Chapter 2 that the file Supermarket Transactions.xlsx contains over 14,000 transactions made by supermarket customers over a period of approximately two years. For this problem, consider this data set the population of transactions.
a. If you were interesting in estimating the mean of Revenue for the population, why might it make sense to use a stratified sample, stratified by product family, to estimate this mean?
b. Suppose you want to generate a stratified random sample, stratified by product family, and have the total sample size be 250. If you use proportional sample sizes, how many transactions should you sample from each of the three product families?
c. Calculate the population standard deviations for each of the three product families. Given these and the discussion in the book, do you think the optimal sample sizes would be much different from the proportional sample sizes?
d. Using the sample sizes from part b, generate a corresponding stratified random sample. What are the individual sample means from the three product families? What are the sample standard deviations?

Sales0
Views75