R Programming

Tutorial for the integration of the software R with introductory statistics
c Grethe Hystad
Copyright
Chapter 3
The Normal Distribution
In this chapter we will discuss the following topics:
The normal density and the density curve with the R-function named dnorm (d ensity).
The cumulative normal distribution function with the R-function named pnorm (probability).
The Quantiles with the R-function named qnorm (quantile).
There are three arguments to the functions dnorm(x, , ) and pnorm(x, , ), where x is
an observation from a normal distribution that has mean and standard deviation . The
argument p to the function qnorm(p, , ) is the proportion of observations in the normal
distribution that are less than or equal to the corresponding quantile x.
The density for a continuous distribution measures the probability of getting a value
close to x. Continuous random variables have a density at a point since they have no
probability at a single point, P (X = x). For the normal distribution we compute the
density using the function dnorm(x, , ).
The function pnorm(x, , ) computes the proportion of observations in the normal
distribution that are less than or equal to x; that is P (X x), where X is N (, ).
The function pnorm(x, , , lower.tail = F ALSE) computes the proportion of observations in the normal distribution that are greater than or equal to x; that is P (X x),
where X is N (, ).
The function qnorm(p, , ) returns the quantile for which there is a probability of p
of getting a value less than or equal to it. Thus, the quantile is the value x such that
P (X x) = p for a given p. I other words, qnorm converts proportions to quantiles
while pnorm converts quantiles to proportions which means that qnorm and pnorm
are inverse functions of each other.
Figure 1: Normal density curve illustrating P (X x) = p
The Normal Density Curve

Problem. Plot the bell curve of a normal distribution with mean 0 and standard deviation
1.5 over the domain 5 x 5.
Solution. Notice that the left and right tail of the distribution will be close to zero for this
choice of domain.
> x=seq(-5,5,0.1)
> plot(x,dnorm(x,0,1.5),type="l",col="green")
Explanation. The code can be explained as follows:

The function x=seq(-5,5,0.1) generates an array of equally spaced points from 5 to
5 in steps of 0.1 assigned x.
The function dnorm(x,0,1.5) computes the density at x for a normal distribution with
mean zero and standard deviation 1.5.
The entry
type="l" (Notice that this is the letter l)
in the function plot connects the points by lines. By default, R plots the points.
The Normal Cumulative Distribution function

The probability that x is in the interval between a and b is the area under the density curve
between a and b.
Problem. Suppose X is normal with mean 527 and standard deviation 105. Compute
P (X 310).
Solution. We want to find the proportion of observations in the distribution that are less
than or equal to 310. That is, the area under the curve to the left of the x-value 310. This
can be done as:
> pnorm(310,527,105)
[1] 0.01938279
Thus, P (X 310) = 0.019.
Problem. The amount of monsoon rain in Tucson is approximately Normally distributed
with mean 5.89 inches and standard deviation 2.23 inches. (Data from 1895-2013) [1].
In what percents of all years is the monsoon rainfall in Tucson between 4 inches and 7 inches?
Solution. Let X be the amount of rain in inches. Here we want to find P (4 X 7). This
can be written as P (X 7) P (X 4), where X is N (5.89, 2.23). In R this can be done as:
> pnorm(7,5.89,2.23)-pnorm(4,5.89,2.23)
[1] 0.4923238
Thus, in 49.23% of all years is the monsoon rainfall in Tucson between 4 inches and 7 inches.
Quantiles for the Normal distribution

Problem. Find the value of x such that the area to its right is 0.1 under the Normal curve
with a mean of 400 and a standard deviation of 83.
Solution. Finding the value of x such that the area to its right is 0.1 is equivalent to finding
the value of x such that the area to its left is 0.9. Thus, we wish to find x such that
P (X x) = 0.9, where X is from N (400, 83). It can be done as:
> qnorm(0.9,400,83)
[1] 506.3688
Hence, x = 506.37 so P (X 506.37) = 0.9 or P (X 506.37) = 0.10.
Notice that this can also be done in R the following way:
> qnorm(0.1,400,83,lower.tail=FALSE)
[1] 506.3688
Explanation. The code can be explained as follows:
The use of the option lower.tail=FALSE in the qnorm function returns the quantile
for which the area p = 0.1 under the normal curve is to the right for the quantile.
If X is standard normal, we can use the functions, dnorm(x), pnorm(x), and qnorm(x),
where the default in R is = 0 and = 1.
If we want to compute the quantiles given in the table of the standard normal distribution,
we will use the function qnorm(x).
Problem. Compute the 1. quartile of the standard normal distribution.
3
Solution. Here we want to find the value of z such that the area to the left of z is 0.25.
That is, we wish to find z such that P (Z z) = 0.25, where Z is standard normal:
> qnorm(0.25)
[1] -0.6744898
Thus, the 1. quartile is 0.674.
Problem. The math SAT scores among U.S. college students is approximately normally
distributed with a mean of 500 and standard deviation of 100. Alf scored 600? What was
his percentile? (The percentile is the value for which a specified proportion of observations
given in percents fall below it.)
Solution. Let X be the SAT score for the student Alf. We want to find P (X 600). Thus,
in R we obtain:
> pnorm(600,500,100)
[1] 0.8413447
Hence, Alfs SAT score is the 84.13 percentile.
Alternatively, we can first standardize the score such that z =
compute P (Z 1.00), where Z is standard normal. We obtain:
600500
100
> pnorm(1)
[1] 0.8413447
References
[1]
National Weather Service Forecast for Tucson, AZ, at

http://www.wrh.noaa.gov/twc/monsoon/monsoon.php
00
= 1.00 and then

R Programming

Uploaded by

Copyright:

Available Formats

R Programming

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

R Programming

Uploaded by

Copyright:

Available Formats

Tutorial for the integration of the software R with introductory statistics

Figure 1: Normal density curve illustrating P (X x) = p

The Normal Density Curve

Explanation. The code can be explained as follows:

The Normal Cumulative Distribution function

Quantiles for the Normal distribution

National Weather Service Forecast for Tucson, AZ, at

= 1.00 and then

You might also like