Vector autoregressions: lag order uncertainty and least absolute deviations

Vector autoregressive models are often used to model multiple time series at once. They are applied in fields such as climatology, biology and economy. This dissertation studies: 1) the influence of the number of lags included in the model; and 2) a new parameter estimation method. We have concluded there is no preferred method to determine the number of lags. The new estimation method is extremely flexible and more efficient if the data shows extreme values.