Skip to main content

STICERD Econometrics Seminar Series

Estimation and Inference with a (Nearly) Singular Jacobian

Adam McCloskey (Brown University), joint with Sukjin Han

Thursday 22 September 2016 14:00 - 15:30

Graham Wallace Room, 5th Floor, Old Building, Houghton Street, London, WC2A 2AE

Many of our seminars and public events this year will continue as online seminars or as online and in person. Please check our website listings and Twitter feed @STICERD_LSE for updates.

Unless otherwise specified, current restrictions mean in-person seminars are only open to members of the LSE community (those with a valid LSE ID card).

Those unable to join the seminars in-person are welcome to participate via zoom.


About this event

This paper develops extremum estimation and inference results for nonlinear models with very general forms of potential identification failure when the source of this identification failure is known. We examine models that may have a general deficient rank Jacobian in certain parts of the parameter space. When identification fails in one of these models, it becomes under-identified and the identification status of individual parameters is not generally straightforward to characterize. We provide a systematic reparameterization procedure that leads to a reparameterized model with straightforward identification status. Using this reparameterization, we determine the asymptotic behavior of standard extremum estimators and Wald statistics under a comprehensive class of parameter sequences characterizing the strength of identification of the model parameters, ranging from non-identification to strong identification. Using the asymptotic results, we propose hypothesis testing methods that make use of a standard Wald statistic and data-dependent critical values, leading to tests with correct asymptotic size regardless of identification strength and good power properties. Importantly, this allows one to directly conduct uniform inference on low-dimensional functions of the model parameters, including one-dimensional subvectors. The paper illustrates these results in three examples: a sample selection model, a triangular threshold crossing model and a collective model for household expenditures.

STICERD Econometrics seminars are held on Thursdays in term time at 14.00-15.30, ONLINE, unless specified otherwise.

Seminar organisers: Professor Tai Otsu and Dr. Vassilis Hajivassiliou.

For further information please contact Lubala Chibwe, either by email: l.chibwe@lse.ac.uk.

Please use this link to subscribe or unsubscribe to STICERD Econometrics mailing list (emetrics).


This event will take place in Graham Wallace Room, 5th Floor, Old Building, Houghton Street, London, WC2A 2AE.