Efficiency in Large Dynamic Panel Models with Common Factor
This paper deals with asymptotically efficient estimation in exchangeable nonlinear dynamicpanel models with common unobservable factor. These models are especially relevantfor applications to large portfolios of credits, corporate bonds, or life insurance contracts, andare recommended in the current regulation in Finance (Basel II and Basel III) and Insurance(Solvency II). The specification accounts for both micro- and macro-dynamics, induced bythe lagged individual observation and the common stochastic factor, respectively. For largecross-sectional and time dimensions n and T, respectively, we derive the efficiency boundand introduce computationally simple efficient estimators for both the micro- and macroparameters.In particular, we show that the fixed effects estimator of the micro-parameteris asymptotically efficient. The results are based on an asymptotic expansion of the loglikelihoodfunction in powers of 1=n. This expansion is used to investigate the second-orderbias properties of the estimators. The results are illustrated with the stochastic migrationmodel for credit risk analysis.
|Date of creation:||2010|
|Contact details of provider:|| Postal: Bâtiment ENSAE, 5 rue Henry LE Chatelier, 91120 Palaiseau|
Phone: 01 41 17 60 81
Web page: http://www.crest.fr
More information through EDIRC