Stat 443: Forecasting
Midterm - October 24th, 2013
4:005:20pm
$0 \X¥\O (\3
(Family name) (Given name)
Name (Print):
Section 001: Rezas class UW Student ID Number:
Aids: Calculator, English to other language dictionary
QUESTION llMARK
|_|1 /2 1
[ "/|2
to
Regression in Forecasting
Sums of squares: AoV
Prediction in Regression
Forecasting and Regression
Part I: Review of simple and multiple linear regression and
prediction interval
STAT 443: Forecasting (Reza Ramezan)
Regression in Forecasting
Sums of squar
Examples
Examples for Sample ACF b(h)
All the R codes for these slides are available in the le
SampleACF.R
Sample ACF
Examples
iid noise
The data is 100 realizations of iid noise N(0, 1).
X
3
1 0
1
2
iid noise
0
20
40
60
80
100
Index
0.6
0.2
0.2
ACF
1.0
s
STAT 443: Assignment 3
(Winter 2014)
SOLUTIONS (Total = 80 marks)
This assignment is due in class on Thursday April 03. For the data analysis section,
you should hand in the R code and output, as well as your interpretations of the outputs.
You will NOT r
Examples
Examples for Sample ACF (h)
All the R codes for these slides are available in the le
SampleACF.R
Sample ACF
Examples
iid noise
The data is 100 realizations of iid noise N (0, 1).
X
3
1 0
1
2
iid noise
0
20
40
60
80
100
Index
0.6
0.2
0.2
ACF
1.0
s
Introduction
Trend Estimation Elimination
Estimation/ Elimination of trend and season
Testing the estimated noise sequence
Smoothing Methods (part I)
Estimating/Eliminating Trend and Seasonality
Trend and Seasonality
Introduction
Trend Estimation Eliminat
Model selection
Making Predictions
Some Tests Based on Residuals
Forecasting and Regression
Part II: Model selection and diagnostics
STAT 443: Forecasting
Model selection
Making Predictions
Some Tests Based on Residuals
Some Model Selection Methods
R 2 an
STAT 443: Assignment 2
(Winter 2014)
SOLUTIONS (Total : 50 marks)
This assignment is due in class on Thursday March 13. For the data analysis section,
you should hand in the R. code and output, as well as your interpretations of the outputs.
You will NOT
Holt-Winters Algorithm
Holt-Winters Algorithm
Holt-Winters method
This generalises exponential smoothing to the case where
there is a trend and seasonality
Following Chateld and Yar (1988) dene trend as
long-term change in the mean level per unit time
Hav
Stat 431
ASSIGNMENT 3
Due: July 2nd, 2013
Reminder:
Your assignment must be handed in by 11:30 on the due date in DWE 3522.
Be sure to include all R code and relevant output for all questions of this assignment.
1. Suppose a sample of insects are divide
Application: Methods of Suicide
Suppose we are interested in examining the nature of the
association among the 3 variables sex, age group, and method of
suicide, in the following table.
Sex
Male
Male
Male
Female
Female
Female
Age( yrs)
10-40
40-70
> 70
10
Stat 431
ASSIGNMENT 3 SOLUTIONS
1. (a) Given the tolerance distribution, the probability of response the dose x is
x
(x) =
exp(u )/)
du
(1 + exp(u )/)2
exp(x )/)
1 + exp(x )/)
1
= + x
=
log
(x)
1 (x)
This implies that it is most appropriate to choose a
Winter 2011
Stat 340
STAT 340
Notes: Please do not expect this test to be representative of what you will see.
1. Consider the function f (x) =
MX (t)
=
R
x2R
=
1
2
=
1
2
=
1
2
=
1
2
=
1
2
=
1
2
=
R
R
R
R
1
2
exp ( jxj) for x 2 R: Determine the moment gen