We are given a total of 3,000,000 people buy lottery tickets for this game and each ticket costs $20. The layout for the prizes is as follows:

Prize | # Winners

20 | 522,000

40 | 261,000

50 | 195,000

500 | 4,000

10,000 | 300

1,000,000 | 3

  1. Calculate the the pmf of p = amount of prize for random ticket
  2. If I purchase one ticket what is my expected gain?
  3. If I pay an extra $10 my prize is multiplied by 2. What is my expected gain if buy a ticket and pay an extra 10?

Solution: This is what I have so far

For the pmf I have p(20) = 522,000/3,000,000 = .174 p(40) = 261,000/3,000,000 = .087 p(50) = .065 p(500) = .00133 p(10,000) = .0001 p(1,000,000)=.000001

To calculate expected gain I know i have to add up the probabilities of winning each of the prizes less the cost. So the expected gain of the $20 prize would be 0(.174) because the cost of the ticket is 20 and so on... thus the expected gain would be:

= 0(.174)+20(.087)+30(.065)+480(.00133)+9980(.0001)+999980(.000001) Is that correct?

The question continues to ask:

Suppose that I continue to buy tickets until I win a prize (of any amount). After I win a prize, I will not buy any more tickets. Let T be the number of tickets that I will buy.

  1. Find the pmf of T if the tickets are not independent of each other
  2. Under the assumption of independence, what is the distribution of T? Write the pmf. How many tickets would I expect to buy?

Any suggestions for these last two questions?

  • $\begingroup$ "Question continues to ask:..." should begin another post, with a link back to this one. $\endgroup$ Feb 22 '17 at 2:05

Mostly.   You forgot to include "not winning any prize", which still costs the ticket price so has a very probable negative gain.

  • $\begingroup$ how to I include this loss in the equation? $\endgroup$
    – Numb3ers
    Feb 22 '17 at 0:53
  • $\begingroup$ $+(-20)(\tfrac{982,303}{3,000,000})$ or simply don't subtract the ticket price until after weighting all the prizes. $\endgroup$ Feb 22 '17 at 0:55
  • 1
    $\begingroup$ It's easier to calculate $\mathbb{E}[\text{Gain}]=\mathbb{E}[\text{Winnings}]-20$, first because you won't forget your 0, then because knowing $\mathbb{E}[\text{Winnings}]$ independent of price is always useful. $\endgroup$ Feb 22 '17 at 1:00
  • $\begingroup$ Precisely so. $\ddot\smile$ $\endgroup$ Feb 22 '17 at 1:03
  • $\begingroup$ so it would be E= 20(.174) + 40(.087) + 50 (.065) + 500(.00133)+ 10000(.0001) + 1,000,000(.000001) - 20 $\endgroup$
    – Numb3ers
    Feb 22 '17 at 1:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.