Holt-Winters: a forecast with trend and seasonality, error on held-out periods
From a series of sales by period builds a forecast with level, trend and seasonal indices (additive or multiplicative), picks the smoothing parameters, tests the accuracy on the last held-out periods (WAPE, bias, a comparison with the naive seasonal forecast) and shows an approximate band.
At a glance
- What question does it answer?
- What sales will be in the coming periods if the seasonal cycle repeats and the trend holds, and whether the model beats the simple “as in the last season”.
- What data is needed?
- A series of sales by period in order, at least two full seasons.
- How to read the result?
- Example on the default values (36 months): the forecast for 12 periods is 17,919 units, the next month 1,203. On the 6 held-out months WAPE is 2.2% against 10.2% for the naive seasonal forecast; the trend is 11.6 per period. Example on the default values: Forecast for 12 periods, total: 17,919; Forecast for the next period: 1,203; WAPE on held-out periods: 2.2%; Bias on held-out periods: −0.2%. The figures are illustrative: enter your own to see your result.
Why and how to reason
A sales series consists of three parts: level (where we are now), trend (where we are heading) and seasonality (how the month affects sales). The Holt-Winters method tracks all three and updates them with each new period, giving more weight to recent data. The level is updated as α × (actual ÷ index) + (1 − α) × (level + trend), the trend as β × (change in level) + (1 − β) × previous trend, the seasonal index as γ × (actual ÷ level) + (1 − γ) × previous. The forecast h periods ahead equals (level + h × trend) × the index of the period. The calculation picks the parameters itself (α 5%, β 1%, γ 50%) and tests accuracy on held-out periods: WAPE 2.2% against 10.2% for the naive seasonal forecast “same as last year”. Level at the end of the series is 1,420, trend 11.6 per period, seasonal range 1.73.
When to use it
It comes in handy for series with clear seasonality and no sharp breaks: product groups, basic range, store traffic, where there are at least two full seasons. Do not use it for series shorter than two seasons, with rare sales and after radical events (closure, change of assortment). Assumptions: seasonality repeats, trend changes smoothly, data is cleaned of shortages and promotions.
How to read the result
The key number is the error on held-out periods (WAPE 2.2%) and the gain against the naive forecast (8 points). If there is no gain, a complex model is not needed, take the naive one. Bias (minus 0.16%) shows the model does not systematically over- or under-forecast. Parameters when fitted: low α and β mean the series is stable, and a high γ tells of changing seasonality. The seasonal range of 1.73 is the strength of seasonality: it suggests how much an order should grow toward the peak. The approximate interval shows the forecast is not a point. It is worth checking the forecast by eye on a chart and match it with the sales plan.
Common mistakes
Feeding a series with gaps and shortages (the model will learn the shortage as a season), taking the season length not matching the data period, fitting parameters on all data without held-out periods, believing the interval as a guarantee. Add known events (promotions, a shifted season) separately, the model does not know about them.
What data is needed?
- A series of sales by period in order, at least two full seasons.
- The season length and the seasonality type: in fashion usually multiplicative.
- How many last periods to hold out for the accuracy test.
Inputs
| Field | Unit | Value in the example | Meaning |
|---|---|---|---|
| Season length | periods | 12 | How many periods in one seasonal cycle: 12 for months, 52 for weeks, 4 for quarters. At least two full seasons of data are needed. Footwear, accessories and children’s have their own values: take the category data. |
| Seasonality type | Multiplicative: the season grows with sales | In fashion seasonality is usually proportional to the sales level, so multiplicative is taken; additive suits a season swing independent of the level. Multiplicative needs sales above zero in every period. | |
| Smoothing parameters | Pick on a grid | The pick scans a grid of values and takes those with the smallest sum of squared one-step errors on the series; the result is reproducible. Set your own if you know how fast demand changes. | |
| Level smoothing (α) | % | 30 | The higher, the faster the level follows the latest sales. Used if the parameters are set by hand. The example value is illustrative. |
| Trend smoothing (β) | % | 10 | The higher, the faster the trend estimate changes; the trend is usually changed slowly. Used with manual parameters. The example value is illustrative. |
| Seasonal smoothing (γ) | % | 30 | The higher, the faster the seasonal indices are updated. Used with manual parameters. The example value is illustrative. |
| Forecast horizon | periods | 12 | How many periods ahead to forecast. The further, the wider the band. |
| Periods to hold out for the test | periods | 6 | The model is built on the series without these periods, forecasts them and the forecast is compared with the actual. Zero: no test. At least two seasons must remain after the hold-out. |
How it is calculated
| Step | Formula |
|---|---|
| Forecast | (level + h × trend) × the period’s seasonal index |
| Level | α × (actual ÷ index) + (1 − α) × (level + trend) |
| Trend | β × (new level − old level) + (1 − β) × the old trend |
| Season | γ × (actual ÷ (level + trend)) + (1 − γ) × the old index |
| Start | the first season’s level, the trend from the gap between the means of the first two seasons, indices from deviations from that trend |
| Test | WAPE = Σ|actual − forecast| ÷ Σactual on the held-out periods |
Worked numbers
Default values from the app; in the interactive version you enter your own.
- Forecast for 12 periods, total
- 17,919
- Forecast for the next period
- 1,203
- WAPE on held-out periods
- 2.2 %The sum of |actual − forecast| ÷ the sum of actuals; the model has not seen the held-out periods. The colour by 15 and 30% thresholds is illustrative.
- Bias on held-out periods
- −0.2 %Plus: the model over-forecasts, minus: under-forecasts.
- MAPE on held-out periods
- 2.1 %
- WAPE of the naive seasonal forecast
- 10.2 %A forecast “as in the same period of the last season”. If the model is no better than it, the complex model is not needed.
- Gain against the naive, p.p.
- 8
- Level at the end of the series
- 1,420
- Trend per period
- 11.6
- Seasonal swing: the largest index to the smallest
- 1.73×
- Parameter α (level)
- 5 %
- Parameter β (trend)
- 1 %
- Parameter γ (season)
- 50 %
- Actual
- Forecast
- Lower bound
- Upper bound
Data as a table
| Period | Actual | Forecast | Lower bound | Upper bound |
|---|---|---|---|---|
| Jan 2023 | 880 | — | — | — |
| Feb 2023 | 910 | — | — | — |
| Mar 2023 | 1,000 | — | — | — |
| Apr 2023 | 1,000 | — | — | — |
| May 2023 | 930 | — | — | — |
| Jun 2023 | 910 | — | — | — |
| Jul 2023 | 880 | — | — | — |
| Aug 2023 | 950 | — | — | — |
| Sep 2023 | 1,150 | — | — | — |
| Oct 2023 | 1,280 | — | — | — |
| Nov 2023 | 1,380 | — | — | — |
| Dec 2023 | 1,640 | — | — | — |
| Jan 2024 | 960 | — | — | — |
| Feb 2024 | 1,010 | — | — | — |
| Mar 2024 | 1,190 | — | — | — |
| Apr 2024 | 1,110 | — | — | — |
| May 2024 | 1,090 | — | — | — |
| Jun 2024 | 1,050 | — | — | — |
| Jul 2024 | 940 | — | — | — |
| Aug 2024 | 1,110 | — | — | — |
| Sep 2024 | 1,310 | — | — | — |
| Oct 2024 | 1,430 | — | — | — |
| Nov 2024 | 1,630 | — | — | — |
| Dec 2024 | 1,750 | — | — | — |
| Jan 2025 | 1,070 | — | — | — |
| Feb 2025 | 1,190 | — | — | — |
| Mar 2025 | 1,300 | — | — | — |
| Apr 2025 | 1,280 | — | — | — |
| May 2025 | 1,220 | — | — | — |
| Jun 2025 | 1,100 | — | — | — |
| Jul 2025 | 1,100 | — | — | — |
| Aug 2025 | 1,240 | — | — | — |
| Sep 2025 | 1,450 | — | — | — |
| Oct 2025 | 1,650 | — | — | — |
| Nov 2025 | 1,710 | — | — | — |
| Dec 2025 | 1,950 | 1,950 | — | — |
| P+1 | — | 1,203 | 1,143 | 1,263 |
| P+2 | — | 1,295 | 1,210 | 1,379 |
| P+3 | — | 1,440 | 1,336 | 1,544 |
| P+4 | — | 1,401 | 1,281 | 1,521 |
| P+5 | — | 1,334 | 1,200 | 1,468 |
| P+6 | — | 1,246 | 1,099 | 1,392 |
| P+7 | — | 1,201 | 1,042 | 1,359 |
| P+8 | — | 1,352 | 1,183 | 1,522 |
| P+9 | — | 1,595 | 1,415 | 1,775 |
| P+10 | — | 1,783 | 1,594 | 1,973 |
| P+11 | — | 1,908 | 1,709 | 2,106 |
| P+12 | — | 2,162 | 1,955 | 2,370 |
Data as a table
| Measure | Index |
|---|---|
| 1 | 84 |
| 2 | 90 |
| 3 | 99 |
| 4 | 96 |
| 5 | 90 |
| 6 | 84 |
| 7 | 80 |
| 8 | 89 |
| 9 | 105 |
| 10 | 116 |
| 11 | 123 |
| 12 | 139 |
Forecast by period
| Period | Forecast | Lower bound 80% | Upper bound 80% |
|---|---|---|---|
| P+1 | 1,203 | 1,143 | 1,263 |
| P+2 | 1,295 | 1,210 | 1,379 |
| P+3 | 1,440 | 1,336 | 1,544 |
| P+4 | 1,401 | 1,281 | 1,521 |
| P+5 | 1,334 | 1,200 | 1,468 |
| P+6 | 1,246 | 1,099 | 1,392 |
| P+7 | 1,201 | 1,042 | 1,359 |
| P+8 | 1,352 | 1,183 | 1,522 |
| P+9 | 1,595 | 1,415 | 1,775 |
| P+10 | 1,783 | 1,594 | 1,973 |
| P+11 | 1,908 | 1,709 | 2,106 |
| P+12 | 2,162 | 1,955 | 2,370 |
A series of 36 periods, season 12. The parameters are picked on a grid over the whole series; the error on the held-out periods uses parameters picked only on the series before them. The band is approximate (80%): from the spread of one-step errors, growing with the root of the horizon.
For new models without history, and periods with stock-outs and advertised promotions the series is distorted: clean it first. Footwear, accessories and children’s have their own season and length. Rare sizes with zeros are counted by the intermittent demand.
Nuances and limits
- Leaving periods with stock-outs and promotions in the series: the season is distorted.
- Judging the model by its error on the very data it was fitted on: a test on held-out periods is needed.
- Forecasting a new style so: there is no series, take analogs.