Actuarial Outpost Weekly Practice Questions - S Broverman
 Register Blogs Wiki FAQ Calendar Search Today's Posts Mark Forums Read
 FlashChat Actuarial Discussion Preliminary Exams CAS/SOA Exams Cyberchat Around the World Suggestions

 Not looking for a job? Tell us about your ideal job, and we'll only contact you when it opens up. https://www.dwsimpson.com/register

 Probability Old Exam P Forum

#131
05-11-2008, 10:04 AM
 victorandjenny Member CAS Join Date: Jan 2008 Posts: 113

Dear Sam,
I'd like to thank you first for your effort of training us. I benefited a lot although I started solving your problems today.
I found that my answer of this question is different from your solution. The process is the same, just the final step of calculation is different.
Best wishes.
Victor

(0.5-1/e)/root of (2/e-eEXP-2)=0.17
N=square of (2.33/0.17)=187.85.
So my answer is N should be at least 188.

Quote:
 Originally Posted by sam_broverman The question can be found at http://www.sambroverman.com/pqw/apr21-08p.pdf -------------------------------------------------------------------------------- Sam Broverman 2brove@rogers.com www.sambroverman.com

Last edited by victorandjenny; 05-11-2008 at 04:28 PM..
#132
05-11-2008, 04:27 PM
 victorandjenny Member CAS Join Date: Jan 2008 Posts: 113

Dear Sam,
Since I am not a native speaker, I was confused by this question. Did they each put one dollar in the pot, or each put \$0.5 only. Surely I got the right answer, because there's no answer for the latter understanding.
Regards.
Victor

Quote:
 Originally Posted by sam_broverman You can find the question at: http://www.sambroverman.com/pqw/mar19-07p.pdf Sam Broverman 2brove@rogers.com
#133
05-11-2008, 07:54 PM
 victorandjenny Member CAS Join Date: Jan 2008 Posts: 113
A question for Professor Broverman

Dear Professor Broverman,
I have a question for this problem.
When we are trying to find out the value of c, we integrate f(x) over x to get 1. However, should we consider f(0)=c, which is not included in the two parts of the integration (-indefinite, 0) and (0, +indefinite)? If yes, the c=1/3 but not 1/2.
Victor

Quote:
 Originally Posted by sam_broverman You can find the question at: http://www.sambroverman.com/mar13p.pdf Sam Broverman 2brove@rogers.com www.sambroverman.com
#134
05-11-2008, 09:19 PM
 Gandalf Site Supporter Site Supporter SOA Join Date: Nov 2001 Location: Middle Earth Posts: 31,442

Quote:
 Originally Posted by victorandjenny Dear Professor Broverman, I have a question for this problem. When we are trying to find out the value of c, we integrate f(x) over x to get 1. However, should we consider f(0)=c, which is not included in the two parts of the integration (-indefinite, 0) and (0, +indefinite)? If yes, the c=1/3 but not 1/2. Thank you in advance. Victor
For a continuous distribution, like this one, there is density but no probability at a point.

The integration could be considered as (-infinity, 0] and (0,+infinity) or (-infinity,0) and [0,+infinity]. The integrals give the same result either way.
#135
05-11-2008, 09:49 PM
 victorandjenny Member CAS Join Date: Jan 2008 Posts: 113

Dear Gandalf,
Thank you very much for your clarification. I understand it now.
Best wishes.
Victor

Quote:
 Originally Posted by Gandalf For a continuous distribution, like this one, there is density but no probability at a point. The integration could be considered as (-infinity, 0] and (0,+infinity) or (-infinity,0) and [0,+infinity]. The integrals give the same result either way.
#136
05-12-2008, 09:20 AM
 sam_broverman Member Join Date: Feb 2005 Posts: 639

Gandalf,

Thank you for answering the question.

victorandjenny,

For the Mar 19/07 question, they both start out
each putting \$1 in the pot. Then more is added
by the person who tosses the dice first.

For the Apr 21/08 question, it appears that I
made an error. I will post a correction.

SB
#137
05-12-2008, 09:20 AM
 sam_broverman Member Join Date: Feb 2005 Posts: 639
May 12, 2008 Exam P Question

The question can be found at
http://www.sambroverman.com/pqw/may12-08p.pdf

--------------------------------------------------------------------------------

Sam Broverman

2brove@rogers.com
www.sambroverman.com
#138
05-12-2008, 04:20 PM
 victorandjenny Member CAS Join Date: Jan 2008 Posts: 113

Dear Professor Broverman,
Thank you very much for your clarification.
I am suing your manual for SOA P (Actex) and I really like your book.
Best wishes.
Victor

Quote:
 Originally Posted by sam_broverman Gandalf, Thank you for answering the question. victorandjenny, For the Mar 19/07 question, they both start out each putting \$1 in the pot. Then more is added by the person who tosses the dice first. For the Apr 21/08 question, it appears that I made an error. I will post a correction. SB
#139
07-11-2008, 01:07 PM
 gametime50 Member Join Date: May 2008 Posts: 82
may 12/08 question

Dear Professor Broverman,
I was trying to solve this problem and could not find how or why it turned out the way it did. Here is the part that i did not understand.

var(E(X|Y))={2=E(X|Y=0) prob 1/3 =(2-4)^2 (1/3)(2/3) = 8/9
{4=E(X|Y=1) prob 2/3

I do not understand the solution to that part or how it was arrived at. Thank you for any help.
#140
07-11-2008, 03:28 PM
 Actuarialsuck Member Join Date: Sep 2007 Posts: 6,156

I believe that's the Bernoulli shortcut?
__________________
Quote:
 Originally Posted by Buru Buru i'm not. i do not troll.