FORECAST ACCURACY

By: Agung Utama

INTRODUCTION
A forecast is never completely accurate, there will
be always deviation from the actual demand.
This difference between the foracast and the actual
is the forecast error.
Although forecast eror is inevitable, the objective of
forecasting is that it be as slight as possible.
There are different measures of forecast error,
including: Mean Absolute Deviation (MAD), Mean
Absolute Percent Deviation (MAPD), Cummulative
Error (CE), and Error Bias (E).
MEAN ABSOLUTE DEVIATION (MAD)
MAD is an average of the difference between the forecast and
actual demand, as computed by the following formula:

MAD= Dt-Ft I
n
Where:
t =The period number
Dt= Demand in period t
Ft = The forecast for period t
n = The total number of periods
I I = Absolute value
The smaller the value of MAD, the more accurate the forecast,
although viewed alone, MAD is difficult to assess.

COMPUTATIONAL VALUES FOR MAD
period Demand
(Dt)
Forecast
Ft ( = 0.30)
Error (et)
(Dt-Ft)
I dt-Ft I
1 37 37.00 -
2 40 37.00 3.00 3.00
3 41 37.90 3.10 3.10
4 37 38.83 -1.83 1.83
5 45 38.28 6.72 6.72
6 50 40.29 9.69 9.69
7 43 43.20 -0.20 0.20
8 47 43.14 3.86 3.86
9 56 44.30 11.70 11.70
10 52 47.81 4.19 4.19
11 55 49.06 5.94 5.94
12 54 50.84 3.15 3.15
557 49.32 53.38
Using the data in the table, MAD is computed:
MAD=I Dt-Ft I
n
= 53l39
11
= 4.85
THE MEAN ABSOLUTE PERCENT DEVIATION
Measures the absolute error as a percentage of
demand rather than per period.
As a result, it eliminates the problem of interpreting the
measure of accuracy relative to the magnitude of the
demand and forecast values, as MAD does.
A lower percent deviation implies a more accurate
forecast.
MAPD = I Dt-Ft I
Dt
= 53.39
520
= 0.10 or 10%
CUMULATIVE ERROR
Cumulative error is computed simply by summing the
forecast errors, as shown in the formula:

E= e
t


A large positive value indicates that the forecast is
probably consistently lower than the actual demand, or
is biased low.
The comulative eror based on the previuos data is
simply computed as: E= e
t

= 49.31



AVERAGE ERROR (BIAS)
It is computed by averaging the comulative error
over the number of time periods
The comulative error is interpreted similarly to the
comulative error. A positive value indicates low bias,
and a negative value indicates high bias. A value
close to zero implies a lack of bias.
The formula is: = e
t
n
= 49.32
11
= 4.48
DISCUSSION QUESTIONS
Registration number for a marketing seminar over
the past 10 weeks are shown below:




a) Starting with week 2 and ending with week 11,
forecast registrations using the nave forecasting
method.
b) Starting with week 3 and ending with week 11,
forecast registrations using a two-week moving
average.


Week 1 2 3 4 5 6 7 8 9 10
Registrations 22 21 25 27 35 29 33 37 41 37
Demand for heart transplant surgery at RSCM has
increased steadily in the past few years:




The director of medical services predicted 6 years ago
that demand in year 1 would be 41 surgeries.
a) Use exponential smoothing, first with a smoothing
constant of 0.6 and then with 1.9, to develop forecasts
for years 2 through 6.
b) Use MAD criterion, which of the two forecasting
methods is best?

Year 1 2 3 4 5 6
Heart
Transplant
45 50 52 56 58 ?
Given the following data, use least squares regression
to derive a trend equation. What is your estimate of the
demand in period 7? In period 12?

Period 1 2 3 4 5 6
Number 7 9 5 11 10 15