Levinson recursion

Levinson recursion, also known as Levinson-Durbin recursion, is an efficient algorithm used to solve the problem of linear prediction in time series analysis, particularly in the context of autoregressive (AR) modeling. The algorithm is named after the mathematicians Norman Levinson and Richard Durbin, who contributed to its development. The primary goal of Levinson recursion is to recursively compute the coefficients of a linear predictor for a stationary time series, which minimizes the prediction error.