Scribd
Upload
102 views

Chapter 4 Time Series Regression Word PDF

The chapter discusses time series regression models for trend. It describes using polynomial functions of time such as linear, quadratic, and polynomial trends to model trends in a time series. It discusses assumptions for regression analysis including that the error terms must be independent, have constant variance, and be normally distributed. It provides an example of using the Durbin-Watson test to check if error terms are autocorrelated.

Uploaded by

shuting_teoh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
102 views

Chapter 4 Time Series Regression Word PDF

The chapter discusses time series regression models for trend. It describes using polynomial functions of time such as linear, quadratic, and polynomial trends to model trends in a time series. It discusses assumptions for regression analysis including that the error terms must be independent, have constant variance, and be normally distributed. It provides an example of using the Durbin-Watson test to check if error terms are autocorrelated.

Uploaded by

shuting_teoh
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Chapter 4: Time Series Regression

Chapter 4:
4.1

Time Series Regression

Modelling Trend by Using Polynomial Functions


A trend model sometimes can be used to describe a time series, y t
The model is defined as followed
yt Tt t
where
y t - the value of the time series in period t
Tt - the trend in time period t
t - the error term in time period t
Use polynomial functions of time to model trend. Some useful trends are
No trend
the model is given by Tt b0
there is no long-run growth or decline in the time series over time
Linear trend
the model is given by Tt b0 b1t
there is straight-line long-run growth ( b1 0 ) or decline ( b1 0 )
over time
Quadratic trend
the model is given by Tt b0 b1t b2 t 2
there is a quadratic (or curvilinear) long-run change over time
Polynomial trend
the model is given by Tt b0 b1t b2 t 2 b p t p

Example 1
Four different time series regression models are being use to analysis the past
24 yearly sales (in thousands pairs) of a shoes company. The summary of
SPSS output for each model is given in Figure 1, 2, 3 and 4. Based on the
given results,
a) Write down the equation of each model.
b) Which model is the best model for forecasting purposes? Explain.
c) Forecast the one year ahead of sales on your best model.

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Figure 1

Figure 2

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Figure 3

Figure 4

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Solution:

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

4.2

Assumptions Underlying Regression Analysis


Error term, t must satisfy
Constant variance
Independence
Normality assumptions (normally distributed)
t ~NID 0, 2

The normality assumption can be checked by constructing histograms,


stem-and-leaf diagrams, normal probability plots and normality test of the
residuals.
To test the independence assumption, check the autocorrelation of the
error terms.
Plot of residual versus time
Positive and
negative
autocorrelation

Little or no
autocorrelation
in error terms:
random pattern

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Durbin-Watson Test
o Durbin-Watson statistics
n

(e
t 2

et 1 ) 2

e
t 1

2
t

o (first-order) positive autocorrelation


hypotheses
H 0 : The error terms are not positively autocorrelated
H1 : The error terms are positively autocorrelated
decision making
- If d d L, , we reject H 0 , i.e. the error terms are positively
autocorrelated
- If d dU , , we fail to reject H 0 , i.e. the error terms are not
positively autocorrelated
- If d L, d dU , , the test is inconclusive
o (first-order) negative autocorrelation
hypotheses
H 0 : The error terms are not negatively autocorrelated

H1 : The error terms are negatively autocorrelated


decision making
- If (4 d ) d L, , we reject H 0 , i.e. the error terms are
negatively autocorrelated
- If (4 d ) dU , , we fail to reject H 0 , i.e. the error terms
are not negatively autocorrelated
- If d L, (4 d ) dU , , the test is inconclusive

o (first-order) positive or negative autocorrelation


hypotheses
H 0 : The error terms are not autocorrelated
H1 : The error terms are positively or negatively
autocorrelated
decision making
- If d d L, or (4 d ) d L, , we reject H 0 , i.e. the error
2

terms are positively or negatively autocorrelated


If d dU , and (4 d ) dU , , we fail to reject H 0 , i.e. the
2

error terms are not autocorrelated


If d L, d dU , or d L, (4 d ) dU , , the test is
2

inconclusive

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Example 2
Given the data the past 24 yearly sales (in thousands pairs) of a shoes
company and the fitted linear regression line, are the error terms
autocorrelated at 98% significance level?

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

Sales,
10
8
15
14
18
20
29
38
35
47
56
70
80
96
98
126
137
158
189
178
201
199
185
205

Solution:

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Example 3
Figure 5 shows the linear regression analysis using SPSS for 24-years sales of
a shoes company. Is the model adequate? Use = 0.02.
Solution:

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Figure 5

SQQS 3033 Business Forecasting

Chapter 4: Time Series Regression

Table 1: Durbin-Watson Table (first-order autocorrelation)


n
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
45
50
55
60
65
70
75
80
85
90
95
100

SQQS 3033 Business Forecasting

dL,0.05
1.08
1.10
1.13
1.16
1.18
1.20
1.22
1.24
1.26
1.27
1.29
1.30
1.32
1.33
1.34
1.35
1.36
1.37
1.38
1.39
1.40
1.41
1.42
1.43
1.43
1.44
1.48
1.50
1.53
1.55
1.57
1.58
2.60
1.61
1.62
1.63
1.64
1.65

dU,0.05
1.36
1.37
1.38
1.39
1.40
1.41
1.42
1.43
1.44
1.45
1.45
1.46
1.47
1.48
1.48
1.49
1.50
1.50
1.51
1.51
1.52
1.52
1.53
1.54
1.54
1.54
1.57
1.59
1.60
1.62
1.63
1.64
1.65
1.66
1.67
1.68
1.69
1.69

dL,0.01
0.81
0.84
0.87
0.90
0.93
0.95
0.97
1.00
1.02
1.04
1.05
1.07
1.09
1.10
1.12
1.13
1.15
1.16
1.17
1.18
1.19
1.21
1.22
1.23
1.24
1.25
1.29
1.32
1.36
1.38
1.41
1.43
1.45
1.47
1.48
1.50
1.51
1.52

dU,0.01
1.07
1.09
1.10
1.12
1.13
1.15
1.16
1.17
1.19
1.20
1.21
1.22
1.23
1.24
1.25
1.26
1.27
1.28
1.29
1.30
1.31
1.32
1.32
1.33
1.34
1.34
1.38
1.40
1.43
1.45
1.47
1.49
1.50
1.52
1.53
1.54
1.55
1.56

10

You might also like