Scribd
Upload
8 views3 pages

MC Techniques: - Inverse Transform Method Cumulative Distribution Function

The document discusses Monte Carlo techniques for generating random numbers from probability distributions. It describes the inverse transform method which uses the cumulative distribution function to generate values. It also covers the acceptance-rejection technique which generates candidate values from an overestimate of the desired distribution and accepts them with a probability equal to the ratio of the desired distribution to the overestimate.

Uploaded by

khushidadhania84
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
8 views3 pages

MC Techniques: - Inverse Transform Method Cumulative Distribution Function

The document discusses Monte Carlo techniques for generating random numbers from probability distributions. It describes the inverse transform method which uses the cumulative distribution function to generate values. It also covers the acceptance-rejection technique which generates candidate values from an overestimate of the desired distribution and accepts them with a probability equal to the ratio of the desired distribution to the overestimate.

Uploaded by

khushidadhania84
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

MC techniques

Suggested readings:
http://pdg.lbl.gov/2005/reviews/monterpp.pdf
• MCs assume a random number generator which generates uniform statistically
independent values in the interval [0,1)
•Inverse Transform Method: if the probability density function is f(x) with -∞<x<∞ its
cumulative distribution function expressing the probability that x ≤a is given by

Notice this should be normalized to 1


If a is chosen with pdf f(a) then the integrated probability up to point a, F(a) is itself a
random variable which will occur with uniform probability in [0,1]. We can find a
unique x chosen from the pdf f for a given u if we set

Provided we can find an inverse function defined by


Inverse Transform Method
For a discrete function
Acceptance-rejection technique
When F(x) is unknown or difficult to work out the inverse transform method cannot
be applied. Let’s suppose that we know enough f(x) so that we can enclose it in a
shape h(x) (easily generated) that is C>1 times f(x). h(x) can be a normalized sum of
uniform distributions. To generate f(x), first we generate a candidate x according to
h(x) and then calculate f(x) and Ch(x). We generate randomly u and test if
uCh(x) ≤ f(x). If so we accept x, if not we reject x and retry.
The efficiency is the ratio of the areas 1/C.

You might also like