P.A. Tinsley

Linear prediction and estimation methods for regression models with stationary stochastic coefficients

Abstract A linear regression model is proposed in which the coefficient vector is a weakly stationary multivariate stochastic process. The model provides a convinient representation of a general class of nonstationary processes. Prediction and estimation methods are proposed that are linear and relatively easy to compute. The proposed procedures are illustrated by estimation of time-varying GNP multipliers of several macro policy instruments over the period 1891-1970. The results are compatible with theoretical priors and suggest that predictability of policy outcomes depends on the mixture of policy instruments.

Indicator and filter attributes of monetary aggregates: A nit-picking case for disaggregation

Abstract All global measures do conceal information. That is their virtue as well as their vice, and the task of science, in economics as elsewhere, is to find and devise aggregates which retain mostly essential information and discard mainly irrelevant information.

Moving Endpoints and the Internal Consistency of Agents‘ Ex Ante Forecasts

Forecasts by rational agents contain embedded initial and terminal boundary conditions. Standard time series models generate two types of long-run boundary values or steady-state endpoints – fixed endpoints and moving average endpoints. Neither can explain the shifting endpoints implied by postwar movements in the cross-section of forward rate forecasts in the term structure or by post-1979 changes in survey estimates of expected long-run inflation. Multiperiod forecasts by a broader class of moving endpoint time series models provide substantially improved tracking of the historical term structure and generally support the internal consistency of the ex ante long-run expectations of bond traders and survey respondents.

The rational expectations approach to economic modelling

Abstract In the two decades since Muth (1961) advanced the concept of rational expectations, explorations of this conjecture have permeated all major areas of economic inquiry: This paper suggests that conventional formulations of the rational expectations postulate violate the axiomatic basis of modem statistical theory by confounding ‘objective’ and ‘subjective’ notions of probability. It is logically impossible to test the rationality of subjective expectations by comparison with observable frequencies. If a rational expectations conjecture is simply imposed on a model, conditions for identification of the model are more stringent than indicated in earlier literature. An alternative model for aggregation of subjective expectations is proposed.

The short-run volatility of money stock targeting

Abstract The altered allocations of money market volatility obtained by alternative monetary policy procedures are illustrated by stochastic simulations of a staff monthly model. The results indicate the nature of the tradeoff between short-run volatility in the money stock and in the funds rate that is available to money stock targeting procedures.

Asymmetric Adjustments of Price and Output

Asymmetries in price adjustment can reconcile contrasts between rapid price movements in inflationary episodes, consistent with classical theories of flexible pricing, and sluggish price responses in contractions, consistent with Keynesian theories of sticky price adjustments. Nonparametric analysis of SIC two-digit industry data indicates that negative asymmetries are more pronounced for real outputs than for nominal outputs, suggesting reversed positive asymmetries in producer pricing. Pricing decision rules are estimated to distinguish between asymmetries in conditioning shocks and asymmetries in producer responses. Two rational motives for asymmetric pricing are supported.

Effective interest rate policies for price stability

Abstract Recent years have seen a trend among central banks of industrial economies to downgrade operational target ranges for exchange rates or monetary aggregates and to adopt price stability as the nominal anchor of monetary policy. This paper contrasts the effects of four simple interest rate policies on the volatility of final demand prices and output. Policy responses are initiated by movements in one of four alternative nominal indicators - nominal output, the money stock, a commodity price index and the final demand price index itself. A strong rank ordering of expected performances is indicated for the four policies. The reasons for this performance ordering of policies are sufficiently general that it is likely to hold for many empirical models of industrial economies.

ON THE STOCHASTIC CONTROL OF LARGE NONLINEAR ECONOMETRIC MODELS

A major impediment to the development of stochastic control strategies which consider parameter uncertainty for large implicitly defined nonlinear econometric models is the development and operation cost of codes for currently proposed strategies. In this study, the parameters are modeled (in case 1) as independent random variables and (in case 2) as first order autoregressive processes. For many policy problems of interest, the objective function contains only a small subset of the total vector of endogenous variables. This fact facilitates calculation of the covariance matrix of the variables in the objective function as a function of the model, the uncertain parameters, and the controls by Monte Carlo methods.

On the Stochastic Control of Large Nonlinear Econometric Models

Abstract A major impediment to the development of stochastic control strategies which consider parameter uncertainty for large implicitly defined nonlinear econometric models is the development and operation cost of codes for currently proposed strategies. In this study, the parameters are modeled (in case 1) as independent random variables and (in case 2) as first order autoregressive processes. For many policy problems of interest, the objective function contains only a small subset of the total vector of endogenous variables. This fact facilitates calculation of the covariance matrix of the variables in the objective function as a function of the model, the uncertain parameters, and the controls by Monte Carlo methods.