PDA

View Full Version : What do you think ?



mcafeep
02-23-2006, 09:51 AM
I am new to stats so I need help

They have also asked the probability of an individual consumer receiving a prize over the course of a year. For this Probability option we will use the Empirical or relative frequency method. Factors governing the out come are:

- There is no more information available as we are in the planning stage.
- We have no number of customers because the card has not been promoted yet.
- We have no history or existing sources of data on the sweepstakes because none have been performed yet.

In answering this question you must CALCULATE and present a probability or a range of probabilities that an individual consumer will receive a prize during the year based on the “frequency” of 52 purchases per year. Please explain the NATURE OF THE PROBABILITY OR RANGE OF PROBABILITIES presented, i.e. explain what the probability number(s) mean.

Hint: Please note that nothing is said of how these 52 purchases will occur IN TIME throughout the year for the average customer.

You must determine how these 52 purchases are employed in concert with the 1000 sweepstake entrees per sweepstake in determining the probability you will provide for winning a sweepstakes.

The consumer sweepstake will be calculated at:


P (E) Relative frequency of E = Frequency of E
. Numbers is trails of Expectations

A sweepstakes occurs every time 1000 purchases accumulate for the bank. A sweepstakes occurs per 1000 purchases on a “RUNNING BASIS” throughout the year every consumer has a central tendency to make 52 purchases per year based on a “FREQUENCY STUDY “performed earlier by the bank. Calculate the probability of a consumer getting at least one prize within a year is:


N(n) =1000 (RUNNING BASIS TRAILS)

N(F)=52 (FREQUENCY STUDY)

52/1000= .052= 5.2%

P (E) 5%

am I thinking right ?

jerryb
02-23-2006, 11:42 AM
i don't think you have enough information, perhaps if you knew the average number of purchases per day and the st. dev of purchases per day youc ould make a dent in this,

but i will tell you that 5% IS WRONG unless the bank has very few customers during the year (<20), which seems unlikely.

cheers
jerry

JohnM
02-23-2006, 12:07 PM
This problem has been discussed before, and is an example of where you need to make an "educated guess" along with some incomplete data. Will it be exactly right? No. Will it be in the ballpark? Yes, probably.

http://www.talkstats.com/probability/218-probability.html?highlight=prize