MA 214: Applied Statistics
Instructor: Ian Johnston
Regression Analysis
Part III
Where Weve Been
We were discussing.
How to fit a linear regression model with one
covariate.
How to quantify the goodness of fit of the
model.
How to verify the adequacy of t
Using JMP
A few side notes.
If you dont know how to interpret a particular part
of a JMP output, read the documentation about
the procedure you are using on the JMP website.
As a shortcut, you can try searching for the name of the
procedure you are usin
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Categorical Data Analysis
MA 214
Where Weve Been
We were discussing.
How to model a response variable that is
continuous, while the covariates (factors) are
continuous or categorical.
MA 214
2
Wher
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
ANOVA
Part I
MA 214
Where Weve Been
We were discussing.
How to develop statistical models where the
response variable Y is continuous and
covariates are (mostly) continuous.
MA 214
2
Where Were Goi
Solutions to Homework 1
Prakash Balachandran
Department of Mathematics & Statistics
Boston University
Monday, September 19, 2011
1. (DAgostino, # 6) According to the Empirical Rule, approximately:
68% of all observations fall between X s and X + s,
95% of
Solutions to Homework 2
MA115, Chapter 2
2. For histograms, all bins have to be touching each other. If not, all points taken off.
(a) 3 points
Frequency
Relative
Frequency
[110,119]
2
0.08
[120,129]
4
MA 214: Applied Statistics
Instructor: Prakash Balachandran
ANOVA
Part I
MA 214
MA 213 Thursday 10/31/2013
Last Time: Chapter 10: One-way ANOVA
This Time: Chapter 10: One-way ANOVA continued
Reading: Chapter 10
Deliverable #5 (Results of Inference Procedu
A regression analysis is inappropriate
when
A. You have two variables that
are measured on an interval
or ra8o scale
B. You want to make predic8ons
for one variable based on
informa8on about another
variable
C. Th
MA 214 Midterm 2 Review
Here we go again!
For a given x, a confidence interval for
E(y) will always be wider than a
prediction interval for y.
1. True
2. False
Recall: standard error for E(y)
Recall that the standard error for an estimate of
E(y) at a gi
MA 214: Applied Statistics
Instructor: Prakash Balachandran
Nonparametric Inference
MA 214
MA 214 Tuesday 12/03/2013
Last Time: Logistic Regression
This Time: Chapter 14 - Nonparametric Inference
Reading: Chapter 14
Problem Set #10: Due Monday, December 9
MA 214: Applied Statistics
Instructor: Prakash Balachandran
Regression Analysis
Part II
MA 214
MA 213 Thursday 10/17/2013
Last Time: Chapter 11: Simple Linear Regression
This Time: Chapter 11 Review + Clickers
Chapter 12: Multiple Regression
Reading: Chap
A few notes about the relationship between the
Chi-squared distribution, and the Z, t, and F
distributions.
Lets take a short break from regression
and discuss the chi-squared distribution!
Whats so special about the chi-squared
distribution? Why do we ca
MA 214
Applied Statistics
Group Projects
The motivation behind the statistics project is to give you an opportunity to put into practice what
we will study in lectures, studio sessions, and problem sets, by designing, implementing, and
presenting the resu
Department of Mathematics and Statistics
Fall 2014
MA 214
Applied Statistics
TIME & LOCATION: Lecture: TR 12:30 - 2:00 PM, CAS226
Lab/Studio: 1 hour/week, Wednesdays Check your schedule
Discussion: 1 hour / week, Mondays-Check your schedule
PROFESSOR:
Ash
Homework 11 Solution Solutions to SAS Problems
1. The ANOVA Procedure Class Level Information Class Levels Values trt 4 control t15 t40 t50 Number of observations 20 The ANOVA Procedure Dependent Variable: time Sum of Source DF Squares Mean Square F
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
ANOVA
Part II
MA 214
Where Weve Been
We were discussing
.
how to develop a model for a continuous
response variable based on one categorical
covariate (factor). This is called one-way
ANOVA.
MA 214
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Logistic Regression
MA 214
Where we have been.
We were discussing how to analyze a
categorical response variable based
on one categorical covariate (factor).
Analysis involved comparing the
frequ
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Regression Analysis
Part II
MA 214
Where Weve Been
Discussed the meaning of Regression
Model
Introduced model fitting via least square
Learnt how and why we carry out statistical
inference of model
Statistical Modeling
MA 214
1
F-test under ANOVA for simple linear
regression model
MA 214
2
F-test under ANOVA for simple linear
regression model
MA 214
3
Model is a good fit to the data in.
1. All four cases
2. Only for case 1
3. Only for case 2
4. Only
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Hypothesis Testing
MA 214
Where Weve Been
Reviewing
the notion of sampling distribution
point
estimation
error
of estimation
confidence
Interval
MA 214
2
Where Were Going
What happens if instead ou
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Regression Analysis
Part I
MA 214
Where Weve Been
Characterizing populations based on
certain features of the populations
(population parameters), such as
mean, median, standard deviation, etc.
Com
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Regression Analysis
Part III
MA 214
Where Weve Been
We were discussing.
How to fit a linear regression model with one
covariate.
How to quantify the goodness of fit of the
model.
How to verify the
MA 214: Applied Statistics
Instructor: Ashis Gangopadhyay
Hypothesis Testing
MA 214
Where Weve Been
Discussing inferential methods for one
sample data and to describe the true
value of a population parameter via
One
sample confidence interval
One
sample h