Practical Time Series Analysis

Confidence interval

One of the commonly asked questions in forecasting is, What is the confidence interval for estimates? The confidence level in a forecasting model is defined by the alpha parameter in the forecast function. The alpha value 0.05 represents an estimate with 95% confidence, which can be interpreted as the estimates returned by the model have 5% probability of not falling in the defined distribution range. The confidence interval is evaluated as follows:

Here, Z_α is the critical value defined based on alpha. For the alpha value 0.05, the critical value is 1.96. The confidence interval with an alpha value 0.05 for the DJIA dataset ...