Full information maximum likelihood estimation of econometric models, linear and nonlinear in variables, is performed by means of two gradient algorithms, using either the Hessian matrix or a computationally simpler approximation. In the first part of the paper, the behavior of the two methods...

It is purpose of this paper to evidence, in the behaviour of the Mini-DMS model for the French economy, some stochastic properties which may confirm, strengthen or sometimes contradict the results obtained from the standard simulation analysis, which is purely deterministic. In particular, the...

Multipliers are often used for selecting alternative policies in economic planning and forecasting. Particular variables like employment, trade balance, inflation or government budget usually impose constraints on the policy action. Therefore a criterion to be preferred to the raw multiplier...

The evaluation of policy actions by means of a large scale econometric model often begins with the analysis of multipliers. A large value of a multiplier, with the right sign, suggests that the policy instrument should be very effective in moving up or down the given target variable. However,...

The drawbacks of predictors obtained with the usual deterministic solution methods in nonlinear systems of stochastic equations have been widely investigated in the literature. Most of the proposed therapies are based on some estimation of the conditional mean of the endogenous variables in the...

The evaluation of policy actions by means of macroeconomic models often begins with the analysis of multipliers. A rough analysis recommends to use those instruments that exhibit large multipliers. Government budget usually imposes some constraints on the policy action. Insted of the raw...

Through Monte Carlo experiments, this paper compares the performances of different gradient optimization algorithms, when performing full information maximum likelihood (FIML) estimation of econometric models. Different matrices are used (Hessian, outer products matrix, GLS-type matrix, as well...

Five alternative techniques have been applied to measure the degree of uncertainty associated with the forecasts produced by a macro-model of the French economy, the Mini-DMS developed at INSEE. They are bootstrap, analytic simulation on coefficients, Monte Carlo on coefficients, parametric...

With most of the available software packages, estimates of the parameter covariance matrix in a GARCH model are usually obtained from the outer products of the first derivatives of the log-likelihoods (BHHH estimator). However, other estimators could be defined and used, analogous to the...