1 / 11

CS 475/575 Slide Set 7

CS 475/575 Slide Set 7. Random Variates 2 M. Overstreet Spring 2005. Uniform distribution. pdf cdf When x=a, F(x)=0, and C=(-a)/(b-a), so F(x) = (x-a)/(b-a) so r=(x-a)/(b-a) or x=r(b-a)+a where r is U(0,1). Exponential distribution - 1. Assumptions:

lawrence-mckenzie
Télécharger la présentation

CS 475/575 Slide Set 7

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. CS 475/575 Slide Set 7 Random Variates 2 M. Overstreet Spring 2005

  2. Uniform distribution pdf cdf When x=a, F(x)=0, and C=(-a)/(b-a), so F(x) = (x-a)/(b-a) so r=(x-a)/(b-a) or x=r(b-a)+a where r is U(0,1)

  3. Exponential distribution - 1 Assumptions: 1. The probability that an event occurs during is 2. is constant and independent of t. 3. The probability that more than one event occurs in approaches 0 as

  4. Exponential dist. - 2 If X is the length of time between successive events then X is said to have an exponential distribution and its PDF is The CDF is

  5. more exp.

  6. even more exp. Note that since F is expressible in closed form (that is, ), we can find : Since (1-r) is identically distributed to r, we use

  7. Normal Recall: Cannot use inverse CDF technique since we have not closed form representation, but do have at least 2 techniques.

  8. Normal - technique 1 Recall central limit theorem: the probability distribution of the sum of N iid random variates with mean  and variance tends toward a normal distribution with mean N and variance /n.

  9. normal - tech 1 cont. Thus if Since for U(a,b),  =(a+b)/2, so pick k=12 to get Problem: not good beyond

  10. Normal - technique 2 Requires generation of normals in pairs: 1. Generate two U(0,1) random variates, 2. Let 3. Let

  11. using Arena to fit distributions • see text, pp. 156-163 • steps: • open Arena • from Tools, select Input Analyzer • select “File,” “new”, then “use existing” icon. • select fit, then “fit all” or a particular distribution

More Related