Microsoft Office Tutorials and References
In Depth Information
Using NORM. DIST and POISSON. DIST to Determine Probabilities
The Poisson distribution is a discrete distribution and is used only with data
that takes on discrete (integer) values, such as counting items.
A Poisson distribution is not always symmetrical, as is the one shown in
Figure 11-9. Negative X values make no sense in a Poisson distribution. After
all, you can’t have fewer than zero people calling in sick! If the mean is a small
value, the distribution will be skewed, as shown in Figure 11-10 for a Poisson
distribution with a mean of 4.
mean of 4.
Excel’s POISSON.DIST function lets you calculate the probability that a
specified number of events will occur. All you need to know is the mean of the
distribution. This function can calculate the probability two ways:
✓ Cumulative: The probability that between 0 and X events will occur.
✓ Noncumulative: The probability that exactly X events will occur.
The two Poisson graphs shown earlier were for noncumulative probabilities.
Figure 11-11 shows the cumulative Poisson distribution corresponding to
Figure 11-9. You can see from this chart that the cumulative probability of 15
events — the probability of 15 or fewer events occurring — is about 0.15.
What if you want to calculate the probability that more than X events will
occur? Simple! Just calculate the cumulative probability for X and subtract the
result from 1.
The POISSON.DIST function takes three arguments:
✓ The first argument is the number of events that you want to calculate
the probability for. This must be an integer value greater than 0.