Partial current information and signal extraction in a rational expectations macroeconomic model: A computational solution

Previous attempts at modelling current observed endogenous financial variables in a macroeconomic model have concentrated on only one variable — the short-term rate of interest. This paper applies a general search algorithm to a macroeconomic model with an observed interest rate and exchange rate to solve the signal extraction problem. Firstly, the algorithm is tested against a linear model with a known analytical solution. Then, the algorithm is applied to all the observed current endogenous variables in a non-linear rational expectations model of the UK. The informational advantage of applying the signal extraction algorithm is evaluated in terms of the forecasting efficiency of the model.