If I have a distribution mass function p(Y) , and values of $X_0,X_1,...,X_i$, I want to assign values to $X_{i+1},...,X_n$ such that the series of X follows p(Y) distribution where Y is the set of all distinct values of X. Is there any technique to do this?
Creating data set to match a particular distribution
0
$\begingroup$
probability-theory
-
0If this is a discrete distribution, then $p(x)$ gives the probability that a random variable is equal to some value. So then the question is: given the probabilities what values should one assign to the random variables $X_{i+1}, \dots, X_n$? – 2010-11-13
-
0@trevor:yes. you are right. – 2010-11-13
-
0What is $X$ here? Is it the sequence $(X_0,\ldots,X_n)$, interpreted as $n$ samples of a discrete probability distribution? If so, I'm not sure there is a meaningful sense in which a sequence of specified samples follows a given distribution, whereas a different sequence does not. For example, does the sequence $(\mathrm{H,H,H,H,H})$ follow the distribution of a fair coin toss? If so, would this be a satisfactory result for you? – 2010-11-13