The Normal Distribution - Assignment 3 | STATS 13, Assignments of Statistics

Material Type: Assignment; Class: Introduction to Statistical Methods for Life and Health Sciences; Subject: Statistics; University: University of California - Los Angeles; Term: Unknown 1989;

Typology: Assignments

Pre 2010

Uploaded on 08/26/2009

koofers-user-51n
koofers-user-51n 🇺🇸

3

(1)

10 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Homework 3: The normal distribution
4 book questions, some R commands
Due 2/23
This homework begins with reading Sections 3.6, 4.1-4.4 and 4.6 of your textbook.
(This will help with your lab due next week also.) The text spends a fair bit of space
in Section 4.3 explaining how to look up information on a normal table. This is not a
skill we will emphasize in this class. Instead of understanding the mechanics of table
lookup, we would prefer that you understand the relationship between a standard
normal and a normal with any other mean µand standard deviation σ.
Questions: 4.3, 4.4, 4.16, 4.18
Rather than use a table to compute the probabilities the book asks for, we would
prefer you use the functionality in R. Specifically, R provides four functions to
help you with the normal distribution: pnorm,qnorm,dnorm and rnorm. (The “p”
because it returns probabilities; the “q” because it returns quantiles or percentiles
as your book calls them; the “d” because it returns the bell curve or density; and
the “r” because it generates random samples.)
First, suppose we are asked about a normal distribution with with mean µ= 5 and
standard deviation σ= 2. We can find the area under the associated bell curve to
the left of 6 (that is, find the chance that Y6 if the distribution of Yhas a bell
shape centered at µ= 5 with σ= 2) with the command
pnorm(6,m=5,s=2)
If you leave out the arguments mand s, then R will assume you are referring to the
standard normal distribution with mean 0 and standard deviation 1. As you will
see in the book, you can transform the problem above to one involving a standard
normal so that the command
pnorm((6-5)/2)
should give you the same answer. Before computers were so commonplace (or to
help instructors give in-class exams), a lot of emphasis was placed on transforming
the problem to so-called standard units and looking things up in the table. Since
pf3

Partial preview of the text

Download The Normal Distribution - Assignment 3 | STATS 13 and more Assignments Statistics in PDF only on Docsity!

Homework 3: The normal distribution

4 book questions, some R commands

Due 2/

This homework begins with reading Sections 3.6, 4.1-4.4 and 4.6 of your textbook. (This will help with your lab due next week also.) The text spends a fair bit of space in Section 4.3 explaining how to look up information on a normal table. This is not a skill we will emphasize in this class. Instead of understanding the mechanics of table lookup, we would prefer that you understand the relationship between a standard normal and a normal with any other mean μ and standard deviation σ.

Questions: 4.3, 4.4, 4.16, 4.

Rather than use a table to compute the probabilities the book asks for, we would prefer you use the functionality in R. Specifically, R provides four functions to help you with the normal distribution: pnorm, qnorm, dnorm and rnorm. (The “p” because it returns probabilities; the “q” because it returns quantiles – or percentiles as your book calls them; the “d” because it returns the bell curve or density; and the “r” because it generates random samples.)

First, suppose we are asked about a normal distribution with with mean μ = 5 and standard deviation σ = 2. We can find the area under the associated bell curve to the left of 6 (that is, find the chance that Y ≤ 6 if the distribution of Y has a bell shape centered at μ = 5 with σ = 2) with the command

pnorm(6,m=5,s=2)

If you leave out the arguments m and s, then R will assume you are referring to the standard normal distribution with mean 0 and standard deviation 1. As you will see in the book, you can transform the problem above to one involving a standard normal so that the command

pnorm((6-5)/2)

should give you the same answer. Before computers were so commonplace (or to help instructors give in-class exams), a lot of emphasis was placed on transforming the problem to so-called standard units and looking things up in the table. Since

you will do all your homework in R, this is a less important skill.

The function qnorm provides the so-called quantile or percentile function. Contin- uing our mean 5 and standard deviation 2 example, we can find the point below which we have 80% of the area under the curve with the command

qnorm(0.8,m=5,s=2)

Similarly, you can find the lower quartile, the median and the upper quartile (the three points that divide the area under the curve into fourths) with the commands

qnorm(0.25,m=5,s=2) qnorm(0.5,m=5,s=2) qnorm(0.75,m=5,s=2)

Or, you can compute all three at once after having “contatenated” the three values into one object c(0.25,0.5,0.75) (technically, this takes our three values and makes them into a vector).

qnorm(c(0.25,0.5,0.75),m=5,s=2)

You should now have three points in a single vector as your answer. As with pnorm, if you do not specify a mean and standard deviation, the function will assume you mean the standard normal. Using this, you can get the same answer by transforming back from standard units

2*qnorm(0.25)+

(see your text for examples). Next, you can draw the normal curve you are interested in using the command dnorm. Below, we create a sequence (a vector) of 100 numbers ranging from 5-32 = -1 to 5+32 = 11 (the mean of our normal plus or minus three standard deviations – which should cover 99% of the area)

x = seq(-1,11,length=100) y = dnorm(x,m=5,s=2) plot(x,y,type="l")

Recall that the argument type tells R you want a line plot. Finally, you can use rnorm to generate a random sample from the indicated normal distribution.