Stat 350 - Assignment 2
Instructor: Jiguo Cao
It is due at 2:00pm Friday, June 15, 2012.
Please drop your assignment to the assignment-dropping box. It is
located right beside K9510, which is one leve
Stat 350 - Assignment 3
Instructor: Jiguo Cao
It is due at 2:00pm Friday, June 29, 2012.
Please drop your assignment to the assignment-dropping box. It is
located right beside K9510, which is one leve
mea 1M 30/2062/1 [email protected]&
STAT 350 Assignment 3
Instructions: It is due at 5:00pm Tuesday, October 25., 2016. You do not need to
attach the R code for this assignment. Please show your calculation c
STAT 350 Assignment 4
Instructions: It is due at 5:00pm Tuesday, November 1, 2016. Please show your calculation clearly. Make sure to answer each question in complete sentences and show your
calculati
STAT 350 Assignment 3
Instructions: It is due at 5:00pm Tuesday, October 25, 2016. You do not need to
attach the R code for this assignment. Please show your calculation clearly.
Please drop your assi
Chapter 8
Indicator Variables
Linear Regression Analysis 5E
Montgomery, Peck & Vining
1
8.1 The General Concept of Indicator Variables
Qualitative variables also known as
categorical variables. Quali
Stat 350 - Assignment 1
Instructor: Jiguo Cao
It is due at 5:00pm Tuesday, October 4, 2016.
Please drop your assignment to the assignment-dropping box. It is
located right beside K9510, which is one l
STAT 350 Assignment 5
Instructions: Instructions: It is due at 5:00pm Tuesday, November 15, 2016. Please
show your calculation clearly. Make sure to answer each question in complete sentences
and show
Stat 350 - Assignment 2
Instructor: Jiguo Cao
October 5, 2016
It is due at 5pm, Tuesday, October 11, 2016.
Questions
Consider the multiple linear model y = X + , where N ormal(0, 2 I)
1. Derive the le
STAT350 Linear Models in Applied Statistics
Tutorial 11 - Summary 1
Shufei Ge
November 24, 2017
Department of Statistics and Actuarial Science
Ofce hour : 11:30-12:30, Friday, AQ4145
Summary
Introduc
Week 10
(Text Sections 16.3) NOTE: On the midterm Q.3, there was a plot of the residuals vs. tted values. [INSERT PICTURE.] Many of you (incorrectly) said that there was a fan shape in this plot. When
Week 2
(Text Chapter 1,2) Least-Squares Estimation (Ch.1) There are many dierent ways of estimating unknown parameters. One of the simplest is least-squares estimation. Denition: The least-squares est
Week 11
(Text Sections 19.2,23.1,23.2)
Example: Hospital stay data Let Yijk be the average length of stay in the k th hospital in region i with average age group j. We assume that the Yijk s are indep
Chapter 10
Variable Selection and
Model Building
Linear Regression Analysis 5E
Montgomery, Peck & Vining
1
10.1
Introduction
10.1.1 Model-Building Problem
Two conflicting goals in regression model bui
Statistics 350, Assignment 1
2015-09-16
This assignment is due, via the drop box outside K9510, by 11:30am on Tuesday
September 29 (no late submissions accepted). Collaboration is welcome, but you mus
mmgm My; (13: #1534 2 W cm; I
I
In)
I
I
Im
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
WI STAB?) f TMJCOYIQKZ [Mockfmmmj w ,_.-,_,
Lets (Enabler ,0 (egkegm When/x ww -
, _ X; Pveolzcyy/QXPanoforH erOH
Week 5
(Text Chapters 5, 6) Multiple Linear Regression (Ch.6) Normally we have more than one predictor. E.g., in the real estate data set, our original goal was to predict the selling price of a new h
Week 9
(Text Sections 2.9) R2 Denition: The coecient of determination, or R2 value, is dened as R2 = SSR SST O
Since SSR + SSE = SST O and all SS 0,SSR SST O 0 R2 1. Interpretation: R2 is the proporti
Week 8
(Text Sections 2.7, 2.8, 7.1)
Sums of Squares Recall:
n n n
(Yi - Y )2 =
i=1 i=1
^ (Yi - Yi )2 +
i=1
^ (Yi - Y )2
SST O = SSE + SSR NOTE: SSE is actually the SS of the residuals, not the error
Week 3
(Text Chapter 3) Checking Assumption #4: Plot of the Residuals vs. Order It's hard to check the independence assumption! One thing we can do is to plot the ri 's vs. the order in which they wer
Week 6
(Text Chapter 7) Multicollinearity (7.6) Multicollinearity is correlation among the regression coecient estimates (the j s). Typically, such correlation arises because of correlation among the
60
/buestion1 .
Based on the question, we need to have 2 indicator variables, which are 3:1 and x2 and each of
them has a piecewise function:
a: cfw_$1 = 1, if Region2
1 x1 = 0, otherwise
:3 cfw_3:2
STAT350 Linear Models in Applied Statistics
Tutorial 10 - Nonparametric Regression
Shufei Ge
November 17, 2017
Department of Statistics and Actuarial Science
Ofce hour : 11:30-12:30, Friday, AQ4145
No
Question 1
a. 80 is -85.4447, 31 is 2.1280 and 82 is 3.4109.
b. The mean ozone will increase 2.128 parts per billion from 1300 to 1500 hours at
Roosevelt Island when the maximum daily temperature incr
Multiple Linear Regression Model
Tutorial 9
Shufei Ge
November 10, 2017
Contents
1. Model Adequacy Checking
2
2. Detecting and Treatment of Outliers
6
1
1.
Model Adequacy Checking
Assumptions in line