Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homework,
1
Introduction
What happens when we want to estimate parameters and nd a measure of variability of the estimate without assuming an underlying distribution?
Upon completion of this lesson, you should be able to:
Recall Sampling Distributions
Understand
Credit Exam for Exam 1
Create a Minitab macro (or R function) that will compute one of the following:
1. Permutation F test
2. Condence interval for (see page 46)
3. Permutation test on Deviances (known location parameters, see page 54)
4. Permutation tes
STAT464 Homework #5
due Monday, 9/26
In a study to see if cloud seeding aects rainfall, the following samples of rainfall were observed:
Test to see if seeding increases rainfall using the.
a) Wilcoxon sum-rank sum test (you may use the normal approximati
STAT464 Homework #4
due Wednesday, 9/21
1. Consider the following observations for two treatment groups: A vs B.
A
B
9
3
18
15
12
6
(a) Find the permutation distribution for these samples. For each possibility, compute the
dierence in sample means and the
STAT 464 HW2 Solution
1. Consider again the exam score data from Homework 1. R software will be needed for
these questions.
(a) Perform the usual t-test of H0 : = 75.5 versus Ha : > 75.5, assuming the
exam scores come from a normal distribution. Report th
STAT 464 HW1 Solution
1. Let X have a binomial distribution with n = 6 and p = 0.2.
(a) What is P (X = k)? For what values of k is this nonzero?
Solution:
n k
6
nk
P (X = k) =
p (1 p)
=
(0.2)k (0.8)6k
k
k
for k = 0, . . . , 6.
For k = 0, . . . , 6, th
Homework #3
1. Recall the sampled values from Homework 2 #3:
0.86
0.48
4.43
0.87
2.51
Use the Komogorov-Smirnov test statistic to test H0 : F (x) = F0 (x), where F0 (x) = 1 ex
is the cdf of an exponential distribution with mean one. You may use = .20.
The
STAT464 Homework #3
due Wednesday, 9/14
1. Recall the sampled values from Homework 2 #3:
0.86
0.48
4.43
0.87
2.51
Use the Komogorov-Smirnov test statistic to test H0 : F (x) = F0 (x), where F0 (x) = 1 ex
is the cdf of an exponential distribution with mean
STAT464 Homework #1
due Wednesday, 8/31
1. Let X have a binomial distribution with n = 6 and p = .2.
(a) What is P (X = k)? For what values of k is this nonzero?
(b) Find P (1 X < 4).
(c) Use the normal distribution to approximate the probability in part
STAT464 Homework #2
due Wednesday, 9/7
1. Consider again the exam score data from Homework 1. R software will be needed for these
questions.
(a) Perform the usual t-test of H0 : = 75.5 versus Ha : > 75.5, assuming the exam
scores come from a normal distri
STAT464 Homework #6
due Wednesday, 10/12
The data in hw6.dat are samples of waiting times from three bus routes (R, V, and S). The data
can be read into R with the following command:
x = read.table(file.choose(), header=T, sep=" ")
1. Find the ANOVA F sta
STAT464 Homework #5
due Monday, 9/26
In a study to see if cloud seeding affects rainfall, the following samples of rainfall were observed:
seeding: 0.74 0.54 0.95 1.01 0.86 0.59 0.43
no seeding: 0.66 0.44 0.13 0.61 0.26 0.29 0.56
Test to see if seeding in
STAT464 Homework #8
due Wednesday, 10/26
1. For the data in the table below,
a) Obtain the permutation distribution of the mean of the dierences. There are 16 possible permutations. For each one, give the dierence for each pair (treatment 1 minus
treatmen
STAT464 Homework #10
due Monday, 11/7
(Adapted from Higgins, Ex4.12 ) Three varieties of soybeans (1=susceptible, 2=moderately resistant, 3=resistant) were measured for iron deficiency in a randomized complete block design.
Data values are on a 1 to 9 poi
STAT464 Homework #4
due Wednesday, 9/21
1. Consider the following observations for two treatment groups: A vs B.
A
B
9
3
18
15
12
6
(a) Find the permutation distribution for these samples. For each possibility, compute the
dierence in sample means and the
STAT464 Homework #9
due Wednesday, 11/2
(Adapted from Higgins, Ex4.7 ) The data in hay.dat are dry matter contents (in kilograms) of hay
obtained from experimental plots. The experiment was designed as a randomized complete block
with six blocks, and the
1
Tests for the Scale Parameter
The tests we have considered so far in this chapter deal with testing measure of center or of location.
In practice, these tests are used in situations when we are testing if one treatment performs better
than another. Ther
1
Other Two-Way Methods
Last Lesson we learned how to handle paired data and two-way layouts with one observation per
cell. We will discuss other methods that handle paired data. We also will discuss how to handle
two-way layouts with more than one observ
STATISTICS 464: Nonparametric Statistical Methods
Instructor: Tracey (Wrobel) Hammel
Email: Use ANGEL email system only
Email: Please remember to use the ANGEL email system under Communicate tab. Emails
received outside of ANGEL may go unanswered! NOTE: I
To run a macro in Minitab:
1. Click on Editor>Enable Commands
2. Open your dataset. Make sure your data is in the appropriate columns.
3. At the MTB< prompt type 0ilelocation\filename.mac
4. The macro should start running.
Troubleshooting:
-Sometimes the
Hypotheses for the Homework
The most important part of hypothesis testing is setting up the null and alternative hypotheses.
If these hypotheses are incorrect, the rest of your work will also be incorrect. The hardest part is
being able to convert a resea
1
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homewor
1
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homewor
1
Correlation and Trends
So far we have been considering situations where the response is quantitative and the explanatory
variable is categoricalical. For example, suppose we want to compare if the median GPAs (response,
quantitative) are dierent based o
1
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homewor
1
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homewor
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homework,
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homework,
Directions
Every Sunday, a homework assignment is due. You are to either type in your answers in Word or
handwrite your answers and scan them into a .pdf le. You must show all your work (do not just
put an answer). Therefore, if you type up your homework,
STAT464 Homework #7
due Wednesday, 10/19
(Adapted from Higgins, Ex3.6 ) The data in exercise 3.6.dat are the head injury readings for
driver-side crash dummies for various classes of vehicles. The vehicle types are coded as 1=Compact,
2=Heavy, 3=Light, 4=