Defining Confidence Intervals

In addition to calculating a base prediction, the most likely scenario, Advanced Predictions can also calculate additional scenarios—the best case and worst case scenarios, also known as a confidence interval.

Based on confidence intervals, Advanced Predictions generate multiple scenarios of ML Predictions and stores the results and in the scenario you select.

• The confidence intervals in a prediction provide an upper and lower bound for predicted output values.
• For example, using the confidence intervals of 10% (P10) and 90% (P90) provides a range of values known as an 80% confidence interval. The observed value is expected to be lower than the P10 value 10% of the time, and the P90 value is expected to be higher than the observed value 90% of the time.

By generating forecasts at P10 and P90, you can expect the true value to fall between those bounds 80% of the time.

To define confidence intervals:

1. Select whether to generate confidence intervals.
2. Select the prediction interval to use for the prediction. The prediction interval, or confidence interval, defines the best case and worst case values for the prediction. For example, if you select 5% and 95%, then the best case scenario is the 95th percentile of the prediction, the maximum value for the prediction, and the worst case scenario is the 5th percentile of the prediction, the lowest value for the prediction. Select the prediction interval to use for the prediction.
3. For Model estimates (fitted values) for historical data, Best Case, and Worst Case, select members from each dimension to define where to store prediction data for fitted values, best case, and worst case scenarios.