What is autoregression?

Autoregression is defined as a statistical analysis method used for modeling and predicting future values in a time series based on its own previous values. This method operates under the assumption that past data points provide information about future values, making it particularly useful in time series forecasting. By employing lagged values of the same variable (e.g., past price points or sales figures), autoregressive models help identify patterns or trends that may recur, thus facilitating accurate predictions.

In contrast, the other options refer to different forecasting methodologies or concepts. Models that require external variables for forecasting do not rely solely on the past values of the series itself, which is a key characteristic of autoregression. Regression models using only time variables are typically not autoregressive, since they do not leverage the relationships among the past values of a variable, focusing instead on considering time as a predictor. Lastly, techniques that correlate independent variables are not specific to time series forecasting, making those options distinct from the concept of autoregression.