Excel FORECAST.ETS.STAT Function

What is the FORECAST.ETS.STAT function?

FORECAST.ETS.STAT is the statistics component of Excel's time series forecasting family. Its Turkish name is TAHMİN.ETS.İSTAT, and it returns statistics that measure the quality and reliability of forecasts made with the ETS (Exponential Smoothing) algorithm. Use it when you want to see the model's alpha, beta, gamma parameters or metrics like MAPE.

While FORECAST.ETS produces the actual forecast, FORECAST.ETS.STAT lets you numerically evaluate how reliable that forecast is. It's a critical tool in data science and advanced forecasting.

FORECAST.ETS.STAT Syntax

=FORECAST.ETS.STAT(values, timeline, statistic_type, [seasonality], [data_completion], [aggregation])

• values: (Required) The historical observation values used for forecasting.
• timeline: (Required) Dates corresponding to the values (must be evenly spaced).
• statistic_type: (Required) A number from 1 to 8:
  • 1: Alpha (level smoothing parameter)
  • 2: Beta (trend smoothing parameter)
  • 3: Gamma (seasonal smoothing parameter)
  • 4: MASE (Mean Absolute Scaled Error)
  • 5: SMAPE (Symmetric Mean Absolute Percentage Error)
  • 6: MAE (Mean Absolute Error)
  • 7: RMSE (Root Mean Squared Error)
  • 8: Detected season step
• seasonality: (Optional) 0 = auto-detect, 1 = non-seasonal, ≥2 = season length.
• data_completion: (Optional) Handling missing data: 0 = fill with zero, 1 = interpolation (default).
• aggregation: (Optional) Method for aggregating duplicate dates: 1=AVERAGE (default), 2=COUNT, 3=COUNTA, 4=MAX, 5=MIN, 6=MEDIAN, 7=SUM.

What FORECAST.ETS.STAT Does — Practical Use Cases

1. Measuring Forecast Accuracy

To check how reliable a forecast made with FORECAST.ETS is, look at MAPE or RMSE values. Lower values (e.g., MAPE < 10) indicate a healthy forecast.

2. Detecting Seasonality

Not sure if your data is seasonal? =FORECAST.ETS.STAT(sales,dates,8) returns the detected season period. 12 indicates yearly, 7 indicates weekly seasonality.

3. Model Parameter Optimization

Alpha, beta, and gamma are auto-tuned by Excel, but seeing the chosen values helps understand model behavior. High alpha = overly sensitive to recent data; low alpha = more stable forecasts.

4. Confidence Score in BI Reports

When telling a manager "Q1 2026 sales = 1.2M," adding a confidence score is professional. If SMAPE = 5%, you can say the model is 95% reliable.

5. Validation in Academic Research

In scientific papers presenting forecast models, error metrics are expected. FORECAST.ETS.STAT delivers these metrics directly.

Advanced Usage Techniques

Interpreting the Statistic Types

Correctly interpreting error metrics is critical:

• MAPE (5): Percentage error. < 10% excellent, 10–20% good, 20–50% moderate, > 50% poor.
• MAE (6): Absolute error in the same unit. 1000 sales with MAE=50 means avg deviation of 50 units.
• RMSE (7): Square root of mean squared error. More sensitive to large deviations.
• MASE (4): Normalized error vs. naive forecast. < 1 means your model beats naive prediction.

Smoothing Parameter Strategy

Alpha, beta, and gamma range between 0 and 1. Excel auto-optimizes them. But values above 0.95 mean overfitting risk (model too sensitive to recent data); below 0.05 means stagnant model that misses new trends.

Working with Missing Data

Real-world data always has gaps. data_completion=1 (interpolation) is usually the best choice; 0 (fill with zero) distorts trends.

Common Mistakes and How to Fix Them

The most common FORECAST.ETS.STAT error is an unevenly spaced timeline. If dates are 1, 2, 3, 5, 6 (with 4 skipped), you'll get a #NUM! error. The second issue is insufficient data — ETS requires at least 2 seasonal periods; for monthly data, that means 24+ months. Third, providing a statistic_type outside 1–8 raises #VALUE!.

Performance and Version Compatibility

FORECAST.ETS.STAT works in Excel 2016 and all later versions, Microsoft 365, and Excel for Web. Performance depends on data size; 100–1000 points are instant, while 10000+ points may take a few seconds. The ETS algorithm performs parameter optimization, so it's slower than basic statistics functions.

Related Functions

• FORECAST.ETS: Computes the actual forecast.
• FORECAST.ETS.SEASONALITY: Returns the season step length.
• FORECAST.ETS.CONFINT: Returns the confidence interval of the forecast.
• FORECAST.LINEAR: Simple linear forecast (legacy FORECAST).
• TREND: Classic linear forecast function.

Conclusion

FORECAST.ETS.STAT is the "quality control" component of Excel's modern time series forecasting ecosystem. It objectively quantifies how reliable your forecasts are, helps you understand the model's parameters and seasonal structure. Anyone working with data analysis, financial modeling, or business intelligence is expected to use these statistics professionally when presenting forecasts. The ETS algorithm is powerful, but blind trust is dangerous — FORECAST.ETS.STAT is how you audit that trust.