This preview shows pages 1–2. Sign up to view the full content.
This document is titled “Homework 3” but will focus on the Ch. 7, EX 4.
Moving average: For a fourperiod moving average, start with
=
+
+
+
L4 D4 D3 D2 D14
; this is
the forecast for the demand of future periods until
D5
is available. Then the formula is
=
+
+
+
L5 D5 D4 D3 D24
. [I don’t see any real benefit for listing values of D in reverse order,
but they are also arranged that way at
http://en.wikipedia.org/wiki/Moving_average
.] If the
exercise only asked for the forecast of monthly demand for year 6, one could just compute
=…=
=
= ,
+
,
+
,
+
,
=
,
F61
F72
L60 8 000 22 000 20 000 20 0004 17 500
. However, to decide how
good the forecasts are, other computations are useful. Columns A through G make pretty
good sense on the “moving average” worksheet, but instead of =SUM($F$6:F6)/(A64) for
Cell H6 one could used =AVERAGE($F$6:F6). [For cell G6 I tried
=AVERAGE($E$6^ 2:E6^ 2) but it didn’t support exponents.] The rest of the columns make
sense, but the bias was requested and Figure 75 (the “template” for moving average) didn’t
include it so Equation 7.25,
= =
Biasn t 1nEt
is used for =SUM($E$6:E6) in Cell K6; column
K is now used for bias instead of tracking signal, so instead of =SUM($E$6:E6)/H6 one can
use = K6/H6. [It is obvious that
TS
1
must be 1 or 1, but I haven’t figured out why following
values are consecutive integers for moving average, simple exponential smoothing, and
Holt’s model.] “ The vast majority of negative biases and the TS that are less than 6
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview. Sign up
to
access the rest of the document.
 Summer '09
 Guan

Click to edit the document details