STA302/1001: Methods of Data Analysis
Instructor: Fang Yao
Chapter 10: Variable Selection
STA302/1001 Lectures p. 1/17
Variable Selection
also known as model selection
goal: given a set of predictor variables X1 , . . . , Xp , we
want to identify the corr
STA302/1001: Quiz #1, 10:2011:00am, October 1, 2013
Let xi denote the predictor variable and yi denote the response variable. The simple linear
regression model is given by yi = 0 + 1 xi + ei , i = 1, . . . , n, where the error ei is independently
and ide
STA 302 / 1001 Answers to recommended practice problems from chapter 7 Note: These are brief answers to the problems and many would need more detail in order to receive full marks on a test or exam. 7.2 It is the extra sum of squares over a model that has
STA 302 / 1001 Fall 2014
Term Test Solutions
LAST NAME : _tions_
STUDENT # :
FIRST NAME: _Solu_
_
STA 302
ENROLLED IN (tick one):
STA 1001
INSTRUCTIONS:
Time: 100 minutes
Aids allowed: calculator
A t-distribution table is provided on the last page
T
2
SCATTERPLOTS AND REGRESSION
values (xi , yi ), i = 1, . . . , n, of (X, Y ) observed on each of n units or cases. In
any particular problem, both X and Y will have other names such as Temperature
or Concentration that are more descriptive of the data th
UNIVERSITY OF TORONTO
Faculty of Arts and Science
DECEMBER EXAMINATIONS 2009
STA 302 H1F / STA 1001 HF
Duration - 3 hours
Aids Allowed: Calculator
LAST NAME:
SOLUTIONS
FIRST NAME:
STUDENT NUMBER:
There are 23 pages including this page.
The last page is
STA 302 / 1001
Lecture 2
1
Simple Linear Regression
Yi =
0 + 1 X i + i
Yi is the response value (random variable)
Xi is the predictor value (known and constant)
0 is the Y-intercept (constant parameter)
1 is the slope (constant parameter)
i is the error
STA 302 / 1001
Lecture 5
1
Residuals Main Results
= 2 (1 )
Proved
= 2
2
Residuals Main Results
, = 2
3
Residuals Main Results
4
Residuals Main Results
, =
1 1
5
Normality of Errors
=
6
Normality of Errors
Residuals can look like they come from a
n
STA 302 / 1001
Answers to recommended practice problems from chapter 3
Note: These are brief answers to the problems and many would need more
detail in order to receive full marks on a test or exam.
3.1 Skip (1).
(2) There are two distinctions to be made.
STA 302 / 1001 Answers to recommended practice problems from chapter 4 Note: These are brief answers to the problems and many would need more detail in order to receive full marks on a test or exam. 4.1 No and no. At least 90% of the time the joint conden
STA 302 / 1001 Answers to recommended practice problems from chapter 6 Note: These are brief answers to the problems and many would need more detail in order to receive full marks on a test or exam. 6.2 (a) Skip this one. There is no intercept and we didn
`2 VC 3 sD ) 2 # v9 3 sD ) 2 sD 6 3 sD ) 2 F ) Q # ec&$5cyTEAVETEAPePm&57 c x r p v q p sl xr n v w q Q Q q ~ |
E
5 ~ f | P z z ~ 5 5 ma | z z x n m n P
cfw_ ~ cfw_ Vcfw_ cfw_ | z z
Vcfw_ ~ Ycfw_ Ycfw_ ub | cfw_ z z y c
n P v wu q p q Yy
n l
Understanding Diagnostic Plots for Linear Regression Analysis | University of Virginia Library Research Data Services + Sciences
Research Data Services + Sciences Home
(http:/data.library.virginia.edu/)
2016-11-14, 10:47 PM
Reference: http:/data.library.v
A2: Analysis to Forced Expiratory Volume data
Last name: LastName
First name: FirstName
Student ID: 00000000
Course section: STA302H1F-L0101
Due: 11pm, NOV 17, 2016
Q1: (5+5=10 pts) Fit a linear model to original data
Q1-a: Scatter plot and residual plot
A1: Brain Size and Intelligence
Last name:
First name:
Student ID: 000000
Course section: STA302/1001H1F-All sections
October 13th, 2016
Q1: t-test for MRIcount between high and low intellegince groups
The null hypothesis assume we have equal means of MRI
UNIVERSITY OF TORONTO
Faculty of Arts and Science
DECEMBER 2003 EXAMINATIONS
STA 302 H1F / 1001 H1F
Duration - 3 hours
Aids Allowed: Calculator
NAME:
SOLUTIONS
STUDENT NUMBER:
There are 22 pages including this page.
The last page is a table of formulae
UNIVERSITY OF TORONTO
Faculty of Arts and Science
DECEMBER EXAMINATIONS 2007
STA 302 H1F / STA 1001 HF
Duration - 3 hours
Aids Allowed: Calculator
LAST NAME:
SOLUTIONS
FIRST NAME:
STUDENT NUMBER:
There are 20 pages including this page.
The last page is
STA 302 H1F / 1001 HF Fall 2009 Test October 22, 2009
LAST NAME:
SOLUTIONS
FIRST NAME:
STUDENT NUMBER: ENROLLED IN: (circle one) STA 302 STA 1001
INSTRUCTIONS: Time: 90 minutes Aids allowed: calculator. A table of values from the t distribution is on the
University of Toronto at Mississauga
STA331H5F 2011
Term Test #5
Aids Allowed: non-graphing calculator without a text keypad
Aids Provided: none
1. (2 marks) Consider the SLR model Y = 0 + 1X + . We often use an F statistic to test the
hypothesis Ho: 1 =
University
University of Toronto at Mississauga
STA331H5F 2011
Term Test #4
Aids Allowed: non-graphing calculator without a text keypad
Aids Provided: none
1. (6 marks) Consider these functions of random variables Y1, Y2, and Y3.
W1 = 2Y1 Y2 + Y3
W2= Y1 Y
University of Toronto at Mississauga
Regression Analysis
STA331H5F 2010
Term Test #4
Aids Allowed: non-graphing calculator without a text keypad, computer output
Aids Provided: none
1. (12 marks) A researcher considers three models:
Model 1
Model 2
Model
University of Toronto at Mississauga
STA331H5F 2011
Term Test #3
Aids Allowed: non-graphing calculator without a text keypad
Aids Provided: none
1. (6 marks) For SLR data, the predictor and the response have both been standardized,
creating
yi* = (yi /) s
University
University of Toronto at Mississauga
Regression Analysis
STA331H5F 2010
Term Test #3
Aids Allowed: non-graphing calculator without a text keypad, computer output
Aids Provided: none
1. (6 marks) Y is a random vector such that
Let W satisfy
a. L
University of Toronto at Mississauga
STA331H5F 2011
Term Test #2
1. (1 mark) Is this statement True or False? If its False, correct it. To correct the
statement, draw a line through the incorrect part and write a correction below.
You are working with a s
University of Toronto at Mississauga
Regression Analysis
STA331H5F 2010
Term Test #2
Aids Allowed: non-graphing calculator without a text keypad
Aids Provided: none
Printed Name: _
Signature: _
Student Number: _
1. (3 marks) The plot to the right is based
University of Toronto at Mississauga
STA331H5F 2011
Term Test #1
VERSION 1
Questions 1-6 are about the following situation An insurance company wants to relate the amount of fire damage (y, in $1,000s) in major
residential fires to the distance between th
