15.

2 FORECAST
ACCURACY
In this section we begin by developing forecasts
for the gasoline time series shown in Table 15.1
using the simplest of all the forecasting
methods, an approach that uses the most
recent week’s sales volume as the forecast for
the next week.

2
 For instance, the
distributor sold 17,000
gallons of gasoline in
Week 1: this value is
used as the forecast
Place your screenshot here for week 2 . Next, we
use 21, the actual
value of sales in Week
2, as the forecast for
week 3, and so on.

3
 The forecasts obtained for the
historical data using this method
are shown in Table 15.7 in the
column labeled Forecast.
Place your screenshot here

Because of it’s simplicity,

this method is often
referred to as Naive
forecasting method.

4
 How accurate are the forecasts
obtained using this naïve forecasting
method?
These measures are used to determine how well a particular forecasting method
is able to reproduce the time series data that are already available. By selecting
the method that is most accurate for the data known we are hope to increase
the likelihood that we will obtain more accurate forecasts for future time
periods.
 The key concept associated with
measuring forecast accuracy is forecast
error.
If we denote Y 1- F as 1 the actual and forecasted values of the time
series for period t, respectively the forecasting error for period t is

6
 “
That is the forecast error for time period t is the difference
between the actual and the forecasted values for period t.

😉
7
For instance, because the distributor actually sold at 21,000 gallons of
gasoline in week 2 and the forecast, using the sales volume in week 1,
was 17,000 gallons, the forecast error in week 2 is

Forecast Error in Week 2 = 21 - 17 = 4

The fact that the forecast error is positive indicates that in
Week 2 the forecasting method underestimated the actual
value of sales.

8
Next, we use 21, the actual value of sales in week 2, as the forecast for
Week 3

Forecast Error in Week 2 = 19 – 21 = -2

In this case, the negative forecast error may be positive or
negative, depending on whether the forecast is to low or high.

9
It is important to note that because we are
using a past value of the time series to
produce a forecast for period t, we do not
have sufficient data to produce a naïve
forecast for the first week of this time series.
A simple measure of forecast accuracy is the mean or average of
the forecast errors. If we have n periods in our time series and k is
the number of periods at the beginning of the time series for
which we cannot produce a naïve forecast, the mean forecast
error (MFE) is

k
Table 15.7 shows that the sum of
the forecast errors for the gasoline
time series is 5; thus the mean or
average error is

5 / 11 = 0.45
Because we do not have
sufficient data to produce a
naïve forecast for the first
week of this time series, we
must adjust our calculations in
both the numerator and
denominator accordingly.
❑ We often use k periods from the time series to produce forecasts,
and so we frequently cannot produce forecasts for the first k
periods.
❑ In those instances the summation in the numerator starts at the first
value of t for which we have produced a forecast (so we begin the
summation (t= k + 1)
❑ The denominator (which is the number of periods in our time series
for which we are able to produce a forecast) will also reflect these
circumstances.
❑ In the example, although the time series consists of 12 values, to compute
the mean error we divided the sum of the forecast errors by 11 because
there are only 11 forecast errors (we cannot generate forecast sales for the
first week using this naïve forecasting method).
❑ Also note that the mean forecast error is positive, which implies that the
method is generally under forecasting; in other words, the observed values
tend to be greater than the forecasted values.
❑ The negative and positive forecast errors tend to offset one another, the
mean error is likely to be small; thus the mean error is not a very useful
measure of forecast accuracy.
 Mean Absolute Error
Is a measure of forecast accuracy that avoids the problem of
positive and negative forecast errors offsetting one another.
MAE is the average of absolute values of the fore cast errors:
 This is also referred to as the
mean absolute deviation or MAD.
Table 15.7 shows that the sum of
absolute values of forecast error is
41; thus
Place your screenshot here

MAE = Average of the absolute value of

forecast errors =
41 / 11 = 3.73
 Mean Squared Error
Is another measure that avoids the problem of positive and
negative errors offsetting each other is obtained by
computing the average of the squared forecast errors:
 From table 15.7 the sum of
the squared errors is 1.79
hence,
Place your screenshot here

MSE = Average of the sum of squared

forecast errors=
179 / 11 = 16.27
The size of MAE and MSE depends upon the scale of the data. As a result, it is difficult
to make comparisons for different time intervals or to make comparisons across
different time series .
To make comparisons such as these we need to work with relative or percentage error
measures.
The mean absolute percentage error, denoted MAPE is such a measure.
 For example, the percentage error corresponding to the
forecast of 17 in Week 2 is computed by dividing the
forecast error in week 2 by the actual value in week 2 and
multiplying the result by 100. For week 2 the percentage
error is computed as follows:
Finally, we find MAPE, which is calculated as:
 Table 15.7 shows that the sum of the absolute MAPE = average of the absolute
values of the percentage errors is 211.69 thus value of percentage forecast errors
= 211.69 =19.24%
11
In summary, using the naïve (most recent observation) forecasting method, we obtained the following measures of
forecast accuracy:

MAE = 3.73
MSE = 16.27
MAPE = 19.24%

These measures of forecast accuracy simply measure how well the forecasting method is able to forecast historical
values of the time series. There is no way to address the issue of accuracy associated with forecasts for future periods.
However, if we select a forecasting method that works well for the historical data, and we have the reason to believe the
historical pattern will continue into the future, we should obtain forecasts that will ultimately be shown to be accurate.
 Let us consider the another method for forecasting the
gasoline sales time series in Table 15.1. We begin by
developing a forecast for Week 2. The forecast for Week
2 is just the time series value in Week 1; thus, the
forecast for Week 2 is 17,000 gallons of gasoline. To
compute the forecast for week 3, we take the average of
the sale values in Weeks 1 and 2. Thus,

4.00

Similarly, the forecast for Week 4 is

23
The forecasts obtained using this
method for the gasoline time series are
shown in Table 15.8 in the column
labeled Forecast. Using the results
shown in Table 15.8, we obtained the
following values of MAE, MSE, and
MAPE:
 We can now compare the accuracy of the two forecasting methods we have considered in this section by
comparing the values of MAE, MSE, and MAPE for each method.

Naïve Method Average of Past Values

MAE 3.73 2.44

MSE 16.27 8.10
MAPE 19.24% 12.85%

For each of these measures, the average of past values provides more accurate forecasts than using the most recent
observation as the forecast for the next period. In general, if the underlying time series is stationary, the average of all
the historical data will provide the most accurate forecast. Measures of forecast accuracy is important factors in
comparing different forecasting methods, but we have to be careful to not rely too heavily upon them. Good judgement
and knowledge about business conditions that might affect the value of the variable to be forecast also have to be
considered carefully when selecting a method. Historical forecast accuracy is not the sole consideration, especially if the
pattern exhibited by the time series is likely to change in the future.

25
Thank you!
Deverly Dawn T. Adrales
Laiah Grace Z. Benavides
(BSA 2 – 2)