Factor Analysis Using Subspace Factor Models: Some Theoretical Results and an Application to UK Inflation Forecasting

Recent work in the macroeconometric literature considers the problem of summarising efficiently a large set of variables and using this summary for a variety of purposes including forecasting. Work in this field has been carried out in a series of recent papers. This paper provides an alternative method for estimating factors derived from a factor state space model. This model has a clear dynamic interpretation. Further, the method does not require iterative estimation techniques and due to a modification introduced, can accommodate cases where the number of variables exceeds the number of observations. The computational cost and robustness of the method is comparable to that of principal component analysis because matrix algebraic methods are used.