Continuous Random Variables
Definition of a Probability Density Function
Since we have all grown up with the concept of probability, there are certain
facts that are already intuitively clear. There cannot be less than a 0
percent chance of something occurring. Also given an experiment, if you
add up the probabilities for all of the possible events the sum should be
P(E) > 0 and SP(E) = 1where the sum is taken over all possible simple events.
This idea work well for discrete random variables, however if the range of outcomes is the real line or an interval, things get more complicated. We call a random variable whose values are all the real numbers or intervals continuous random variables. For continuous random variables, it does not make sense to add up all the probabilities, however the integral extends the idea of integration. This leads us to the key definition.
Remark: It is possible to have the limits be .
f(x) = x + 0.5
be a function of a continuous random variable defined on [0,1]. Show that f(x) is a probability density function.
Clearly, f(x) > 0 on [0,1], hence we need only to check that the integral equals 1.
Find k such that
f(x) = ke-2x
is a probability density function defined on [0, ]. Then find P(0 < x < 0.8)
Since this integral must equal 1, we get
k = 2
The probability density function for the result of asking 1000 people if they think that the president is doing a good job is defined by
Find the probability that more than 54% of 1000 randomly selected people think that the president is doing a good job.
We want to find
P(0< x < 0.54)
We use the definition to get
We use a calculator to get the estimate of 0.135. Hence there is a 13.6% probability that more than 54% of these 1000 people will think that the president is doing a good job.