
Advanced Statistics Advanced Probability and Statistics Math Forum 
 LinkBack  Thread Tools  Display Modes 
July 27th, 2013, 11:41 AM  #1 
Newbie Joined: Jan 2012 Posts: 3 Thanks: 0  Conditional probability distribution change of variables
Hello, I have a simple question regarding changing variables in a conditional distribution. I have two independent variables where r is "rate" (can be any positive real number although most likely to be around 1) and t is "time" (positive integers ie 1,2,3,4...). I have a conditional probability function (really a probability density function) of the form which is "probability that the rate is r (within an interval \mathrm{d}r) at time t" This has a normalization condition which means that there is some rate at any given time I am actually interested in finding a different conditional probability function where R is the cumulative rate up to time t So if if have outcomes for that are then the cumulative rate is , which is to say the product of for i up to Again, this would have to have a normalization condition since for any give time there has be some cumulative rate. If you can help me find from I would greatly appreciate it. Thank you very much for you help. Ilya 
July 27th, 2013, 12:59 PM  #2 
Senior Member Joined: Jun 2013 From: London, England Posts: 1,316 Thanks: 116  Re: Conditional probability distribution change of variables
Hi Ilya, I think that might be very difficult to do for a general pdf P(r). I tried it with P(r) constant between 1/2 and 3/2 and worked out the pdf for R, the product of the first two r1 and r2. It was: P(R) = 2R  1/2 (for R in [1/4, 3/4]) P(R) = 3/2  2R/3 (for R in [3/4, 9/4] You might be able to generalise this for the product of the first n rates for this or other simple pdf's. It would be good, therefore, if you knew what your pdf P(r) is. 
July 27th, 2013, 02:48 PM  #3 
Newbie Joined: Jan 2012 Posts: 3 Thanks: 0  Re: Conditional probability distribution change of variables
The problem is that the pdfs are obtained numerically from empirical data and are not "nice". One thing I can do if I can't find an analytic approach is to do it numerically. I will take N rates which are described by P(r,1) and crossmultiply with N rates described by P(r,2). I will then rebin the resulting rates back to N rates and obtain a pdf (this will be my P(R,2)). I will then crossmultiply that by N rates from P(r,3) and then rebin to find P(R,3) and so on. The problem is that I fear N will have to be quite big for this to work, I need to go for somewhat large times, and I have thousands of different processes (each with their one P's) for which I will need to do this. So all in all it would be quite computationally expensive (for N= 1000, max t=20, and 1000 different processes I would need to do 20 billion multiplications and 20 thousand rebinnings). 

Tags 
change, conditional, distribution, probability, variables 
Search tags for this page 
Click on a term to search for related topics.

Thread Tools  
Display Modes  

Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Change in variables?  Andrew.F  Calculus  0  September 17th, 2012 05:08 PM 
Conditional distribution of xbar given x1  jigsaw  Advanced Statistics  3  August 18th, 2012 12:26 PM 
Joint probability distribution of random variables  meph1st0pheles  Advanced Statistics  1  March 23rd, 2010 05:47 PM 
Conditional Probability and Random Variables  edspecks2004  Advanced Statistics  0  August 18th, 2009 03:11 AM 
Change of variables  tmac3813  Calculus  5  December 17th, 2008 05:39 PM 