3022 HW 4.1 Solution
1. Textbook 4.10
We compute the natural logarithms of the GDP data and store the results in LogGDP. A scatterplot of
LogGDP versus Religiosity is shown below. We see a negative association so
Stat3022 HW6.1 Solution
Problem Set 6.22
The total number of observations is 100 for this problem.Also provide the three
p-values for Face, Gender and their interaction.
The df for both Face and Gender is 1 since each factor has two levels. The interactio
6.2 - Addressing Violated Assumptions Through Transformation | STAT 461
1 of 8
Analysis of Variance
We can often correct non-normality and unequal variances by transforming the response variable. That i
Chap 8: Overview of Experimental Design
School of Statistics, University of Minnesota
1 / 16
To make useful conclusions, randomization is of essential
As we have seen before in the examples, both controlled
3022 HW 5.1 Solution
a. Identify the explanatory and response variables.
The explanatory variable is type of font, and there are four different fonts being used. The response variable
is the final exam score, which is being used as a measure of stude
Put your answers for multiple choice questions here (use capital letters):
Part #2 (30 points total) (Questions 11 - 15)
A certain Minneapolis company is interested in how much of a households garbage contains glas
Ratio estimator: An estimator of the population mean or total based on a ratio with
an auxiliary quantity for which the population mean or total is known.
Regression estimator: An estimator of the population mean or total based on a
1.7 Exercises lg
For Further Reading
The American Statistical Association series What is a Survey? provides an intro-
duction to survey sampling, with examples of many of the concepts discussed in
Chapter 1. In particular, see the chapter Ju
13.3 Chapter Summary
Multiple samples from a population may be used to estimate its size. In the simplest
form, two independent SRSs are taken and the number of population units found
in both SRSs is used to estimate the populatio
#Run the following code line by line.
boxplot(physical.exercise~live.on.campus, outline= F, xlab="live on campus",
ylab = "time spent exercising")
#lab 8 part c
#Do a simulation with 1 sample. We will get 1 "phat" in this case.
#Do a simulation with 100 sample. We will get 100 "phat"s in this case.
for(i in 1:100)cfw_
a. 1 : the interacting with a teller at the bank
2 : the using ATMs
3 : the using banks Internet banking service
Ho: 1 = 2 = 3
Ha: at least two of the population means are different
b. df 1 = 3-1=2
df 2 =400-2=397
F=3, the ri
a. I would find more surprising when flipping the coin 500 times and observing all
b. Because you observed the sequences of flipping coin by doing many times, the
sequences of flips tend to repeat. So the proportions of heads whe
a. point estimate = 0.548
b. interval estimate = (0.548-0.03, 0.548+0.03) =(0.518, 0.578)
c. The difference is that there a point to estimate for the point estimate, but
there is a range for interval to estimate.
a. Null hypothesis Ho: the variables marital happiness and family income
Alternative hypothesis Ha: the variables marital happiness and family
income are dependent
b. df = (r-1)(c-1)= (3-1)(3-1) = 4
2 = 9.4
a. An alternative hypothesis because the parameter in this statement falls in
alternative range of values.
b. A null hypothesis because the parameter in this statement takes a
c. An alternative hypothesis be
Let X = the profit of the farmer. X has the values of $80,000, $50,000 and
$20,000. And the probabilities are 0.7, 0.2, and 0.1 respectively.
P(X 50000) = P(X=50000) + P(X=20000)
a. Sample space S =cfw_Clinton, Sanders, OMalley
b. R delegate from a rural region
B delegate support Bernie Sanders
c. P(R) = 946/1660 = 0.569
d. Rc means the delegates who are not from a rural region.
P(Rc) = 1- P(R) = 1- 0.
a. X = number of years of education for self-employed individuals
in the United States
b. x ~N(13.6,
) = N(13.6, .3)
The mean of the sampling distribution of years of education for a
random sample of size 100 is 13.
Review of Hypothesis Testing
September 12, 2016
Large-sample z-tests for (difference in) proportions
t-tests for (difference in) means
Large-sample z-test for a population proportion
Hypothesis test for a population proportion
Let X1 , . . . , Xn be
October 11, 2016
1. In this problem, you will analyze a dataset. The data are from an experiment where 24
animals were assigned to one of four diets and then blood coagulation times were measured
for each animal. These data are available for