TABLE OF CONTENTS The equiprobable model discussed in the probability primer is quite limiting. How, for instance, could we represent the experiment of throwing a weighted coin which has a probability of 1 / ?2 of coming up heads? We would have to draw from a box containing an infinite number of 0's and 1's. Clearly we need a model allowing different probabilities to be associated with different sample points so let's begin afresh with a more general model. As before, the description of an experiment with unpredictable or stochastic outcomes starts with a list or set of all possible outcomes. We call this set the sample space and denote it by ?. Each outcome is represented by an element of this sample space and this sample point is denoted by ?. Consider the following examples:
An appropriate sample space might be
where h denotes heads and t denotes tails.
Suppose an urn contains two red balls, one white ball and one blue ball and balls are drawn at random without replacement until the blue one is drawn. An appropriatc sample space might be
where the order in which the balls are drawn is indicated by the vector notation.
An appropriate sample space might be
where h denotes heads and t denotes tails.
The sample spaces are...
