transformation method
Summary
The transformation method is a way of sampling from univariate probability distributions by sampling a uniform random variable and inverting the CDF.
Context
This concept has the prerequisites:
- Monte Carlo estimation
- cumulative distribution function (The transformation method involves inverting the CDF.)
Core resources (we're sorry, we haven't finished tracking down resources for this concept yet)
Supplemental resources (the following are optional, but you may find them useful)
-Free-
→ Machine learning summer school: Markov chain Monte Carlo (2009)
-Paid-
→ Pattern Recognition and Machine Learning
A textbook for a graduate machine learning course, with a focus on Bayesian methods.
Location:
Section 11.1.1, pages 526-528
See also
- Other ways of sampling from a 1-D distribution include:
- rejection sampling
- slice sampling , an iterative method
- the polar coordinates trick for sampling from a Gaussian