Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/301118 
Year of Publication: 
2024
Series/Report no.: 
LEM Working Paper Series No. 2024/18
Publisher: 
Scuola Superiore Sant'Anna, Laboratory of Economics and Management (LEM), Pisa
Abstract: 
Linking the statistic and the machine learning literature, we provide new general results on the convergence of stochastic approximation schemes and inexact Newton methods. Building on these results, we put forward a new optimization scheme that we call generalized inexact Newton method (GINM). We extensively discuss the theoretical and the computational aspects of the GINM. The results apply to both deterministic and stochastic approximation schemes, and are particular effective in the case in which the objective function to be optimized is highly irregular and/or the stochastic equicontinuity hypothesis is violated. Examples are common in dynamic discrete choice models and complex simulation models characterized by nonlinearities and high levels of heterogeneity. The theory is supported by extensive Monte Carlo experiments.
Subjects: 
Optimization
stochastic approximation
Newton-Raphson methods
asymptotic convergence
M-estimation
stochastic equicontinuity
JEL: 
C61
C15
C44
Document Type: 
Working Paper

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.