Infinitesimal Robustness for Diffusions

We develop infinitesimally robust statistical procedures for general diffusion pro-cesses. We first prove existence and uniqueness of the times series influence functionof conditionally unbiased M{estimators for ergodic and stationary diffusions, underweak conditions on the (martingale) estimating function used. We then character-ize the robustness of M{estimators for diffusions and derive a class of conditionallyunbiased optimal robust estimators.[...]

MoreLess

Year of publication:	2008-07-01
Authors:	La Vecchia, Davide ; Trojani, Fabio
Institutions:	Finrisk
Subject:	Expansion \| expansion \| Sattelpunktmethode \| Green-Funktion \| Influence Function

freely available

Full text |

More access options

Check Google Scholar

In German libraries (KVK)

I need help

1 Introduction
2 Infinitesimal Robustness for Diffusions
2.1 General Setting
2.2 Infinitesimal Robustness
2.3 Martingale Estimating Functions for Diffusions
3 Optimal Robust Conditionally Unbiased Etimators
3.1 Estimators with Bounded Self{Standardized Sensitivity
3.2 Computation of ¿ (Xi¡1; µ0)
3.3 Algorithm
4 Monte Carlo Simulation and EmpiricalApplication
4.1 Monte Carlo Setting
4.2 Monte Carlo Results
4.3 Real{Data Application
5 Conclusions
APPENDIX
References

Persistent link: https://www.econbiz.de/10005868932