A prediction interval is a range of values associated with a prediction and a confidence level (for example, 95%). For a given set of prediction assumptions, future observations can be expected to fall within the interval a specified percent of the time. For example, if the prediction interval is 10-20, and the confidence level is 95%, you can expect 95% of observations to fall between 10 and 20, and 5% of observations to fall outside that range. The prediction interval takes into account both randomness in the model and uncertainty in the parameter estimates. Prediction intervals are often more useful than a single predicted value because they reflect the uncertainty inherent in predictive modeling.
To view a prediction interval:
The prediction interval will then appear in place of the predicted value.