Lagrößenreihen

Lagrößenreihen, often translated as "lag size series" or "lagged series," is a concept primarily used in time series analysis and econometrics. It refers to a set of variables that are lagged versions of an original time series. For instance, if we have a time series Xt, a lag size series would include Xt-1, Xt-2, Xt-3, and so on, up to a chosen lag k. These lagged values are then used as independent variables in regression models to explain or predict the current value of Xt.

The purpose of incorporating lagged variables is to capture the temporal dependencies within a time series. Many economic and financial phenomena exhibit persistence, meaning that past values influence future values. For example, current consumption might be influenced by consumption in the previous period. Lagrößenreihen allows researchers to quantify these relationships. The selection of the appropriate lag length (k) is crucial and is often determined through statistical methods such as the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC), or by examining autocorrelation and partial autocorrelation functions. Overfitting can occur if too many lags are included, while omitting important lags can lead to biased estimates and poor model performance. Lagrößenreihen are a fundamental building block in the construction of autoregressive (AR) and autoregressive moving-average (ARMA) models, which are widely employed for forecasting and understanding dynamic systems.