Professional Documents
Culture Documents
Cliff T. Ragsdale
Chapter 11
n
Yi Y
i
– the mean square error,
MAD =
– root mean square error.
i 1 n
100 n Yi Ŷi
MAPE =
n i 1 Yi
MSE =
Yi Yi
n
2
i 1 n
RMSE MSE
We will focus on the MSE.
Extrapolation Models
Extrapolation models try to account for the past
behavior of a time series variable in an effort to
predict the future behavior of the variable.
f Y , Y , Y ,
Yt 1 t t 1 t 2
Yt Yt-1 Yt-k +1
Yt 1
k
Y24 Y23 36 + 35
Y25
355
.
2 2
Y
Y 35.5 + 36
Y26
25 24
35.75
2 2
Weighted Moving Average
The moving average technique assigns equal
weight to all previous observations
1 1 1
Yt 1 Yt Yt-1 Yt- k -1
k k k
The weighted moving average technique allows
for different weights to be assigned to previous
observations.
w Y w Y w Y
Yt 1 1 t 2 t-1 k t- k -1
where 0 wi 1 and w
i 1
40
38
Units Sold
36
34
32
Number of VCRs Sold
30 Exp. Smoothing alpha=0.1
Exp. Smoothing alpha=0.9
28
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Time Period
Implementing the Model
Note that,
35.81, for t = 25, 26, 27,
Yt
Seasonality
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Tim e Period
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
Tim e Period
Stationary Data With
Additive Seasonal Effects
Ŷt n E t St n p
where
E t (Yt - St-p ) (1 - )E t 1
St (Yt - E t ) (1 - )St p
0 1
0 1
p represents the number of seasonal periods
Ŷt n E t St n p
where
E t (Yt /St-p ) (1 - )E t 1
St (Yt /E t ) (1 - )St p
0 1
0 1
p represents the number of seasonal periods
Ŷ24 n E 24 S 24 n 4
Ŷ20 n E 20 nT20
Ŷt n E t nTt
where
Et = Yt + (1-)(Et-1+ Tt-1)
Tt = (Et Et-1) + (1-) Tt-1
0 1 and 0 1
Y t b0 b1X1t
where X1t t
For example:
X11 1, X12 2, X13 3,
Implementing the Model
b b X 3751
Y . 92.6255 21 2320.3
21 0 1 121
b b X 3751
Y . 92.6255 22 2412.9
22 0 1 122
b b X 3751
Y . 92.6255 23 2505.6
23 0 1 123
b b X 3751
Y . 92.6255 24 2598.2
24 0 1 124
The TREND() Function
TREND(Y-range, X-range, X-value for prediction)
where:
Y-range is the spreadsheet range containing the dependent
Y variable,
X-range is the spreadsheet range containing the
independent X variable(s),
X-value for prediction is a cell (or cells) containing the
values for the independent X variable(s) for which we want
an estimated value of Y.
b b X b X
Yt 0 1 1t 2 2t