Money demand and seigniorage-maximizing inflation in Chile: approximatin, learning, and estimation with neutral networks

• Autores: Paul D. McNelis
• Localización: Revista de análisis económico, ISSN-e 0718-8870, ISSN 0716-5927, Vol. 13, Nº 2, 1998, págs. 3-24
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • The traditional cash in advance macroeconomic models are characterized by a constant velocity of money. Based on the Lucas and Stokey (1987) model, this paper studies the behavior of velocity and money demand for the U.S., simulating an economy which includes stochastic monetary growth (monetary policy), and income taxes (fiscal policy). The results of the simulations are compared with the actual data using several methods. First, the classical metric of standard errors and correlations are evaluated using block-Wald testing procedures. Next, we implement a well specified VAR estimation to study the impulse response functions of interest rates, velocity and the deficit, among other variables. The impulse responses of the model with both policies (with and without fiscal sector) are compared with the corresponding impulse responses for the data. As a third distance evaluation method, based on Braun (1994), the money demand was studied under the Canonical Cointegrating Regression approach. As a conclusion, and based on the three metrics, the model is not rejected in its ability to reproduce an important proportion of the observed volatility in the U.S.