Scribd
Upload
6K views14 pages

Practice 1 From Introductory Time Series With R

This document summarizes practice exercises from the book Introductory Time Series with R. It includes R code to analyze time series data on housing approvals and economic activity, complaints about a motor organization, and wine sales. Methods demonstrated include plotting, autocorrelation functions, Holt-Winters exponential smoothing with and without trends or seasonality, and Bass model fitting. Results like fitted values, residuals, coefficients and forecasts are examined.

Uploaded by

api-285777244
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
6K views14 pages

Practice 1 From Introductory Time Series With R

This document summarizes practice exercises from the book Introductory Time Series with R. It includes R code to analyze time series data on housing approvals and economic activity, complaints about a motor organization, and wine sales. Methods demonstrated include plotting, autocorrelation functions, Holt-Winters exponential smoothing with and without trends or seasonality, and Bass model fitting. Results like fitted values, residuals, coefficients and forecasts are examined.

Uploaded by

api-285777244
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/ 14

Practice 1 from Introductory Time Series with R

YIK LUN, KEI


[email protected]
This paper is a practice from the book called Introductory Time Series with R
by Cowpertwait, Paul SP, and Andrew V. Metcalfe. All R codes and comments
below are belonged to the book and authors.

6000

10000

14000

www <- "http://elena.aut.ac.nz/~pcowpert/ts/ApprovActiv.dat"


Build.dat <- read.table(www, header=T) ; attach(Build.dat)
App.ts <- ts(Approvals, start = c(1996,1), freq=4)
Act.ts <- ts(Activity, start = c(1996,1), freq=4)
ts.plot(App.ts, Act.ts, lty = c(1,3))

1996

1998

2000

2002
Time

2004

2006

acf(ts.union(App.ts, Act.ts))

0.6
0.2

0.6

App.ts & Act.ts

0.2

ACF

App.ts

0.0 0.5 1.0 1.5 2.0 2.5 3.0

Act.ts & App.ts

Act.ts

0.2

0.6

Lag

0.6

Lag

0.2

ACF

0.0 0.5 1.0 1.5 2.0 2.5 3.0

3.0

2.0

1.0

0.0

0.0 0.5 1.0 1.5 2.0 2.5 3.0

Lag

Lag

app.ran <- decompose(App.ts)$random # generate NAs


app.ran.ts <- window (app.ran, start = c(1996, 3),end=c(2005,2) )
act.ran <- decompose (Act.ts)$random
act.ran.ts <- window (act.ran, start = c(1996, 3),end=c(2005,2) )
acf (ts.union(app.ran.ts, act.ran.ts))

0.5
0.5

0.5

app.ran.ts & act.ran.ts

0.5

ACF

app.ran.ts

0.5

1.0

1.5

2.0

2.5

3.0

0.0

0.5

1.0

1.5

2.0

act.ran.ts & app.ran.ts

act.ran.ts

2.5

3.0

2.5

3.0

0.5

0.5

Lag

0.5

Lag

0.5

ACF

0.0

3.0

2.0

1.0

0.0

0.0

Lag

0.5

1.0

1.5
Lag

ccf (app.ran.ts, act.ran.ts)

2.0

0.2
0.4

0.0

ACF

0.4

0.6

app.ran.ts & act.ran.ts

Lag

T79 <- 1:10


Tdelt <- (1:100) / 10
Sales <- c(840,1470,2110,4000, 7590, 10950, 10530, 9470, 7790, 5890)
Cusales <- cumsum(Sales)
Bass.nls <- nls(Sales ~ M * ( ((P+Q)^2 / P) * exp(-(P+Q) * T79) ) / (1+(Q/P)*exp(-(P+Q)*T79))^2, start =
summary(Bass.nls)
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##

Formula: Sales ~ M * (((P + Q)^2/P) * exp(-(P + Q) * T79))/(1 + (Q/P) *


exp(-(P + Q) * T79))^2
Parameters:
Estimate Std. Error t value Pr(>|t|)
M 6.798e+04 3.128e+03
21.74 1.10e-07
P 6.594e-03 1.430e-03
4.61 0.00245
Q 6.381e-01 4.140e-02
15.41 1.17e-06
--Signif. codes: 0 '***' 0.001 '**' 0.01

***
**
***
'*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 727.2 on 7 degrees of freedom


Number of iterations to convergence: 8
Achieved convergence tolerance: 7.338e-06

6000
2000

Sales per year

10000

Bcoef <- coef(Bass.nls)


m <- Bcoef[1]
p <- Bcoef[2]
q <- Bcoef[3]
ngete <- exp(-(p+q) * Tdelt)
Bpdf <- m * ( (p+q)^2 / p ) * ngete / (1 + (q/p) * ngete)^2
plot(Tdelt, Bpdf, xlab = "Year from 1979",
ylab = "Sales per year", type='l')
points(T79, Sales)

Year from 1979

Bcdf <- m * (1 - ngete)/(1 + (q/p)*ngete)


plot(Tdelt, Bcdf, xlab = "Year from 1979",
ylab = "Cumulative sales", type='l')
points(T79, Cusales)

10

50000
30000
0 10000

Cumulative sales

Year from 1979

www <- "http://elena.aut.ac.nz/~pcowpert/ts/motororg.dat"


Motor.dat <- read.table(www, header = T)
attach(Motor.dat)
Comp.ts <- ts(complaints, start = c(1996, 1), freq = 12)
plot(Comp.ts, xlab = "Time / months", ylab = "Complaints")

10

35
30
25
20
15
5

10

Complaints

1996

1997

1998

1999

2000

Time / months

Comp.hw1 <- HoltWinters(complaints, beta = FALSE, gamma = FALSE)


Comp.hw1
##
##
##
##
##
##
##
##
##
##
##
##
##

Holt-Winters exponential smoothing without trend and without seasonal component.


Call:
HoltWinters(x = complaints, beta = FALSE, gamma = FALSE)
Smoothing parameters:
alpha: 0.1429622
beta : FALSE
gamma: FALSE
Coefficients:
[,1]
a 17.70343

plot(Comp.hw1)

25
20
15
5

10

Observed / Fitted

30

35

HoltWinters filtering

10

20

30

40

Time
Comp.hw1$SSE
## [1] 2502.028
Comp.hw2 <- HoltWinters(complaints, alpha = 0.2, beta=FALSE, gamma=FALSE)
Comp.hw2
##
##
##
##
##
##
##
##
##
##
##
##
##

Holt-Winters exponential smoothing without trend and without seasonal component.


Call:
HoltWinters(x = complaints, alpha = 0.2, beta = FALSE, gamma = FALSE)
Smoothing parameters:
alpha: 0.2
beta : FALSE
gamma: FALSE
Coefficients:
[,1]
a 17.97913

Comp.hw2$SSE
## [1] 2526.39
8

100 200 300 400 500 600

sales (1000 litres)

www <- "http://elena.aut.ac.nz/~pcowpert/ts/wine.dat"


wine.dat <- read.table(www, header = T)
attach (wine.dat)
sweetw.ts <- ts(sweetw, start = c(1980,1), freq = 12)
plot(sweetw.ts, xlab= "Time (months)", ylab = "sales (1000 litres)")

1980

1985

1990

1995

Time (months)
sweetw.hw <- HoltWinters (sweetw.ts, seasonal = "mult")
sweetw.hw
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##

Holt-Winters exponential smoothing with trend and multiplicative seasonal component.


Call:
HoltWinters(x = sweetw.ts, seasonal = "mult")
Smoothing parameters:
alpha: 0.4086698
beta : 0
gamma: 0.4929402
Coefficients:
[,1]
a
285.6890314
b
1.3509615
s1
0.9498541
s2
0.9767623
9

##
##
##
##
##
##
##
##
##
##

s3
s4
s5
s6
s7
s8
s9
s10
s11
s12

1.0275900
1.1991924
1.5463100
0.6730235
0.8925981
0.7557814
0.8227500
0.7241711
0.7434861
0.9472648

sweetw.hw$coef
##
a
## 285.6890314
##
s5
##
1.5463100
##
s11
##
0.7434861

b
1.3509615
s6
0.6730235
s12
0.9472648

s1
0.9498541
s7
0.8925981

s2
0.9767623
s8
0.7557814

sweetw.hw$SSE
## [1] 477693.9
sqrt(sweetw.hw$SSE/length(sweetw))
## [1] 50.54219
sd(sweetw)
## [1] 121.3908
plot (sweetw.hw$fitted)

10

s3
1.0275900
s9
0.8227500

s4
1.1991924
s10
0.7241711

0.8

1.2

season
0.8 1.2 1.6 100

trend
300

level
100 300 500

xhat

sweetw.hw$fitted

1985
1990

Time

plot (sweetw.hw)

11
1995

100 200 300 400 500 600

Observed / Fitted

HoltWinters filtering

1985

1990
Time

data(AirPassengers)
AP <- AirPassengers
AP.hw <- HoltWinters(AP, seasonal = "mult")
plot(AP.hw)

12

1995

500
400
300
100

200

Observed / Fitted

600

HoltWinters filtering

1950

1952

1954

1956
Time

AP.predict <- predict(AP.hw, n.ahead = 4 * 12)


ts.plot(AP, AP.predict, lty = 1:2)

13

1958

1960

700
500
300
100

1950

1955

1960

1965

Time

AP.hw$alpha; AP.hw$beta; AP.hw$gamma


##
alpha
## 0.2755925
##
beta
## 0.03269295
##
gamma
## 0.8707292

Reference:
Cowpertwait, Paul SP, and Andrew V. Metcalfe. Introductory time
series withR. Springer Science & Business Media, 2009.

14

You might also like