THE SPECULATIVE DEMAND FOR MONEY, ENDOGENOUS INFORMATION AND PORTFOLIO ADJUSTMENT (BAYESIAN STATISTICS, DECISION MAKING, TRANSACTION COSTS, RISK).

摘要

This dissertation examines the effects of endogenously determined information on the speculative demand for money and portfolio adjustment over time. Individuals do not possess perfect information, nor can information be acquired costlessly. Therefore, the amount of information must be a choice variable, endogenously determined by the individual.;The microeconomic model is extended to examine the effects of endogenous information on portfolio adjustment in response to a shift in the mean of the return distribution. The individual is found to adjust the portfolio gradually over several periods as information about the new mean is acquired. The speed with which the adjustment occurs depends on the amount of information collected, and therefore, on the determinants of the optimal demand for information. Specifically, anything that causes the optimal amount of information to increase will cause a more rapid adjustment, and vice versa.;The optimal amount of information varies directly with the expected return. Therefore, the higher the expected return, the faster the adjustment. This theoretical prediction was used to respecify the standarad money demand adjustment equation. Specifically, the adjustment coefficient is hypothesized to vary over time, in direct relation to the expected return. Using two different proxies for expected return, the money demand adjustment equation is estimated. The results are consistent with the theoretical hypotheses in most cases. Furthermore, a variable coefficient of adjustment also implies variable short term income and interest rate elasticities, an idea that has significant implications for monetary policy.;The dissertation modifies Tobin's model of speculative demand for money in order to explicitly analyze the effects of information collection on the expectation forming process of the individual. This is done using Bayesian statistical theory. The demand for costly information and the optimal portfolio allocation are derived.