# Finding moment generating functions

1. Mar 20, 2013

### DODGEVIPER13

1. The problem statement, all variables and given/known data

2. Relevant equations

3. The attempt at a solution
Well I believe it to be Binomial as the probabilities never change. So I used it to solve for the moment generating function, I do howvere want to know if what I have done is correct and also how I went from sigma((Pe^t)^y(n y)q^(n-y)) as the book did this?

#### Attached Files:

• ###### EPSON019.jpg
File size:
14.6 KB
Views:
93
2. Mar 20, 2013

### haruspex

Isn't p the probability of one head in one toss? How is it 0.75?

3. Mar 20, 2013

### DODGEVIPER13

Ok so p should be .50 but does my work look ok other than that and the obvious mathematical error due to that on the second part

4. Mar 20, 2013

### DODGEVIPER13

After the final sigma part like I asked above how do I arrive at the moment generating function like I did?

5. Mar 20, 2013

### haruspex

In the third line of part 2, you've written E[Y] = M'(t) by mistake for M'(0).
In the next line, you've left something out when applying the product rule to get M''.

6. Mar 20, 2013

### haruspex

It's just the binomial expansion (run in reverse).

7. Mar 20, 2013

### DODGEVIPER13

Ok a bit lost on what was wrong with m" it looks right to me hmmm where should I look

8. Mar 20, 2013

### haruspex

Looks like I need to take a break. You're quite right, and the rest of your work looks good too.

9. Mar 20, 2013

### DODGEVIPER13

Ok so should I change p to .5 and q to .5 and I wind up with 1.5

10. Mar 20, 2013

### DODGEVIPER13

I changed p from .75 to .5 as you said

11. Mar 20, 2013

### DODGEVIPER13

Also for the first part I now have (.50e^t+.50)^2

12. Mar 20, 2013

### haruspex

All good.

13. Mar 20, 2013

### DODGEVIPER13

Wait why is p what you said should I not also include the probability that you flip a head and head thus making it .75

14. Mar 20, 2013

### DODGEVIPER13

When you say all good was it all good before or all good now?

15. Mar 20, 2013

### haruspex

When you wrote the first two lines of your solution, what were you defining p to be? What do you get if you set n=2 and write out the sum in the second line?
You showed E[Y] = np. If n=2 and p=.75, would you really expect an average of 1.5 heads in two tosses?

16. Mar 20, 2013

### DODGEVIPER13

I was defining p to be the possible heads observed by the 2 flips and to the final question no I would not so I guess what I have changed it to is correct

17. Mar 20, 2013

### haruspex

But p is a probability, not a number of heads. If you wanted to approach it that way you'd need three probabilities: 0, 1 or 2 heads.

18. Mar 20, 2013

### DODGEVIPER13

Ah ok got it now thanks so the way I have it now is correct

19. Mar 20, 2013

Yes.