For numbers outside of the range (0,1), Forecast_ETS_STAT returns an error. A smaller interval implies more confidence in the prediction for this specific point.įor example, for a 90% confidence interval, a 90% confidence level is computed (90% of future points are to fall within this radius from prediction). Using confidence intervals can help you grasp the accuracy of the predicted model. Returns the step size detected in the historical timeline.Ī confidence interval ( Arg4) of 95% means that 95% of future points are expected to fall within this radius from the result that Forecast_ETS forecasted (with normal distribution). Returns the root mean squared error metric, a measure of the differences between predicted and observed values. Returns the symmetric mean absolute percentage error metric, an accuracy measure based on percentage errors. Returns the mean absolute scaled error metric, a measure of the accuracy of forecasts. Returns the trend value parameter-a higher value gives more weight to the recent trend. Returns the base value parameter-a higher value gives more weight to recent data points. The following optional statistics can be returned:
The statistic_type parameter ( Arg3) indicates which statistic is requested by this function. If the ranges of the timeline and values aren't all of the same size, Forecast_ETS_STAT returns run-time error 1004. If the timeline contains duplicate values, Forecast_ETS_STAT also returns an error. If Forecast_ETS_STAT can't identify a constant step in the timeline, it returns run-time error 1004. It'sn't necessary to sort the timeline ( Arg2), because Forecast_ETS_STAT sorts it implicitly for calculations. See Remarks.Īggregation: Although the timeline requires a constant step between data points, Forecast_ETS_STAT aggregates multiple points that have the same time stamp. See Remarks.ĭata completions: Although the timeline requires a constant step between data points, Forecast_ETS_STAT supports up to 30% missing data, and automatically adjusts for it.
Statistic_type: A numeric value between 1 and 8, indicating which statistic will be returned for the calculated forecast.Ĭonfidence level: A numerical value between 0 and 1 (exclusive), indicating a confidence level for the calculated confidence interval. The values in the timeline must have a consistent step between them and can't be zero. Timeline: the independent array or range of dates or numeric data. Values: the historical values, for which you want to forecast the next points. Forecast_ETS_STAT ( Arg1, Arg2, Arg3, Arg4, Arg5, Arg6)Įxpression A variable that represents a WorksheetFunction object.
Returns a statistical value as a result of time series forecasting.
0 kommentar(er)
