Tracking Signal in Forecasting
by Dr Muhammad Al-Salamah
A tracking signal indicates if the forecast is consistently biased high or low. It is computed by dividing the cumulative error by the cumulative mean absolute deviation, or MAD:
The tracking signal is recomputed each period, with updated, "running" values of cumulative error and MAD. The movement of the tracking signal is compared to control limits; as long as the tracking signal is within these limits, the forecast is in control. Control limits of ±2 to ±5 are used most frequently.
To illustrate the way the tracking signal is computed, take the actual sales and forecasts for 6 periods:
|
Period |
Actual |
Forecast |
Deviation |
Cumulative deviation |
Absolute deviation |
Cumulative absolute deviation |
MAD |
Tracking signal |
|
1 |
98 |
95 |
3 |
3 |
3 |
3 |
3 |
1 |
|
2 |
94 |
95 |
-1 |
2 |
1 |
4 |
2 |
-0.5 |
|
3 |
97 |
95 |
2 |
4 |
2 |
6 |
2 |
1 |
|
4 |
93 |
95 |
-2 |
2 |
2 |
8 |
2 |
-1 |
|
5 |
94 |
95 |
-1 |
1 |
1 |
9 |
1.8 |
-0.56 |
|
6 |
90 |
95 |
-5 |
-4 |
5 |
14 |
2.33 |
-2.15 |
The tracking signal can be thought of as a scaled deviation of the forecast from the actual sales figures.
The plot of the tracking signal shows it fluctuates about zero:
If the company uses ±2 control limits, the forecasting technique has to be recalibrated or changed after period 6, as indicated by the tracking signal falling below the lower control limit.