ScholarLens Back to search
1985•Munich Personal RePEc Archive (Ludwig Maximilian University of Munich)Requires access

Gradient methods in FIML estimation of econometric models

Giorgio Calzolari, Lorenzo Panattoni

Open publisher page 0 citations

Abstract

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 as a mixture of them).

Open-access reader

About this research paper

What this paper is about

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 as a mixture of them).

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

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 as a mixture of them).

Key concepts: Hessian matrix, Econometric model, Monte Carlo method, Matrix (chemical analysis), Mathematics, Estimation, Econometrics, Maximum likelihood

Related papers

Related paper 1Related paper 2Related paper 3
Back to paper searchBrowse research topicsOriginal source
Gradient methods in FIML estimation of econometric models — Research Paper | ScholarLens