The dataset in the file $Shopping.csv$ linked below reports information for an "approached in-store at random" sample of 50 of a store's customers. For each sampled customer, both the number of store visits and number of items purchased last year are recorded. The store management is interested in estimating the number of items purchased per customer last year. As you can verify, the average number of items purchased last year amongst the sampled customers is . However, you are concerned that the nature of the sampling is such that frequent customers are over-represented in the sample. To deal with this, you assume that the probability of "catching" a specific customer in the sample is proportional to his/her number of store visits. Using this assumption, give a better estimate of the target than . Report your answer to two decimal places.

LINK for Shopping.csv ( [ the link to 'Download the csv file here' is broken ] )