12.8 Models with Both Quantitative and Qualitative
Variables
November 7, 2016
(Mizzou)
STAT3500
November 7, 2016
1 / 11
Introduction and Motivating Example
So far, we have discussed models with quantitative predictor variables
or qualitative predictor var
12.6 Quadratic and Other Higher-Order Models
October 31, 2016
(Mizzou)
STAT3500
October 31, 2016
1 / 13
Motivation
So far, we have only considered models which allow for straight-line
relationships.
What if y and x have some nonlinear relationship?
y
x
Co
12.9 Comparing Nested Models
November 9, 2016
(Mizzou)
STAT3500
November 9, 2016
1/8
Introduction
So far we have talked about quantitative and qualitative variables,
quadratic terms, interaction terms, and combinations of these.
Our next task is to choose
15.2 The Sign Test
November 30, 2016
(Mizzou)
STAT3500
November 30, 2016
1 / 10
Single Population Inferences: the Sign Test
The sign test is used to make inferences about the central tendency
of a single population
Note that we are interested in the centr
12.7 Qualitative Variable Models
November 2, 2016
(Mizzou)
STAT3500
November 2, 2016
1/7
Dummy Variable
So far, the models weve considered have only included quantitative
variables.
We can also use qualitative variables as predictor variables.
Need to cod
15.1 Distribution-Free Tests
November 30, 2016
(Mizzou)
STAT3500
November 30, 2016
1/4
Distribution-Free Tests
Any test which has a distributional assumption about the response
variable is a parametric test.
Ex: ANOVA requires that the response from each
12.2 The First-Order Model: Estimating and Making
Inferences about the Parameters
October 12, 2016
(Mizzou)
STAT3500
October 12, 2016
1 / 18
Introduction
A first-order model includes only terms for quantitative predictor
variables, and does not include an
12.3 Evaluating Overall Model Utility
October 25, 2016
(Mizzou)
STAT3500
October 25, 2016
1 / 11
Introduction
If we are interested in checking how well the model fits the data, we
could perform t-tests for each parameter, but this would increase our
chanc
12.1 Multiple Regression Models
October 12, 2016
(Mizzou)
STAT3500
October 12, 2016
1/8
Introduction
In the previous chapter, we discussed simple linear regression:
Introduced the straight-line regression model relating a response
variable y to a predicto
11.5 The Coefficients of Correlation and Determination
October 3, 2016
(Mizzou)
STAT3500
October 3, 2016
1 / 12
Linear Relationship
Recall that we are interested in a linear relationship between two
continuous variables, x and y .
In previous sections, we
12.5 Interaction models
October 26, 2016
(Mizzou)
STAT3500
October 26, 2016
1/8
First Order Model
In the previous sections, we have talked about first-order regression
models. For example, consider the first-order model with two
predictor variables:
E (y
11.6 Estimation and Prediction
October 5, 2016
(Mizzou)
STAT3500
October 5, 2016
1/6
Estimation and Prediction
Estimation: estimate the mean value of y , E (y ), given a value of xp
y = 0 + 1 xp
the underlying equation is E (y ) = 0 + 1 xp
Prediction: pre
12.4 Using the Model for Estimation and Prediction
October 26, 2016
(Mizzou)
STAT3500
October 26, 2016
1/3
Estimation and Prediction
Estimation and prediction for multiple linear regression is found
exactly as it was for simple linear regression.
Plug new
9.2 The Completely Randomized Design(CRD):
Single Factor
September 1, 2016
(Mizzou)
STAT3500
September 1, 2016
1 / 22
Goal of Chapter 9
Analyze data collected from a designed experiment using statistical
technique called analysis of variance, or ANOVA
(M
11.4 Assessing the Utility of the Model: Making
Inference about the Slope 1
September 28, 2016
(Mizzou)
STAT3500
September 28, 2016
1 / 12
So far, we have looked at
1
2
3
4
The regression model
How to estimate 0 and 1 using method of least squares
The dis
11.3 Model Assumptions
September 28, 2016
(Mizzou)
STAT3500
September 28, 2016
1 / 18
Introduction
Recall that the straight-line model y = 0 + 1 x + .
The regression line is y = 0 + 1 x.
After we estimate the parameters 0 and 1 , we need to specify the
pr
9.4 The Randomized Block Design (RBD)
September 8, 2016
(Mizzou)
STAT3500
September 8, 2016
1 / 14
Motivation
Previously, we considered completely randomized designs.
For each treatment, we selected independent, random samples of
experimental units.
If th
9.1 Elements of a Designed Experiment
August 29, 2016
(Mizzou)
STAT3500
August 29, 2016
1 / 15
Outline
Designed experiment and observational studies
Chapter9.1: Elements of a Designed Experiment
Chapter9.2: Completely Randomized Design: Single Factor
Chap
11.1 Probabilistic Model
October 1, 2016
(Mizzou)
STAT3500
October 1, 2016
1 / 13
What weve done so far
1
2
Methods for estimating and testing population mean and other
parameters for a single sample (Stat 2200/2500)
Extended these methods to allow for co
11.2 Fitting the Model: The Least Squares Approach
October 1, 2016
(Mizzou)
STAT3500
October 1, 2016
1 / 15
Review
Recall the simple linear regression model is
y = 0 + 1 x +
y is the response variable
x is the predictor/independent variable
0 is the y-in
9.3 Multiple Comparisons of Means
September 5, 2016
(Mizzou)
STAT3500
September 5, 2016
1 / 11
Example
Consider workbook problem WB9-3. Three groups of children were
randomly assigned to do trick-or-treat in different areas of a city
(busy, crowded and ru
Problem 8: Assume that the occurrence of tornadoes in a Midwestern town is believed to
be a Poisson process with a rate of 2.7 tornadoes per year.
(a) What is the probability that there will be exactly two tornadoes in a given year?
Let X denote the numbe
Problem 1: A standard deck of cards contains 52 cards, 4 representing each of 13 different
denominations.
(a) Find the number of ways to draw a 5 card poker hand called four of a kind; this hand
contains four cards of the same denomination and one card of
n-
on to predict the number of ma
- e ressi '
Problem 1. (58 pts) A manufacturer of borler drums wants to use r g n several boilers were collected
. - ' data 0 _
hours needed to erect the drums in future projects. To accemiIISh "5" (X1: lb/hr), boilerW
i
Exam 1
This test is worth a total of 100 points. You have to show your work to receive partial or full credit. Five oints will be
deducted from the test score for return it unstapled.
Problem 1. cfw_26 pts) An article in Industrial Quality Control, descri
I
Statistics 3500: Introduction to Statistics II
Exam 2 Chapter 9
Spring 2016 First Name: Last Name:
Exam 2
This test is worth a total of 100 points. You have to show your work to receive partial or full credit. Egg
points will be deducted from the test s
Statistics 3500: Introduction to Statistics II
Exam 3 - Chapter 12
Spring 2016
First Name: _ Last Name: _
Exam 3
This test is worth a total of 100 points. You have to show your work to receive partial or full credit. Five
points will be deducted from the
Statistics 3500: Introduction to Statistics II
Exam 2 - Chapter 9
Spring 2016
First Name: _ Last Name: _
Exam 2
This test is worth a total of 100 points. You have to show your work to receive partial or full credit. Five
points will be deducted from the t