A prediction interval is a statistical range that is used to estimate the likely value of a single future observation based on a fitted model. It provides an interval that is expected to contain the actual value of that future observation with a specified level of confidence (e.g., 95% confidence).
New to topics? Read the docs here!