Abstract
In this paper, we analyze properties of the Continuous Updating Estimator (CUE) proposed by Hansen et al. (1996), which has been suggested as a solution to the finite sample bias problems of the two-step GMM estimator. We show that the estimator should be expected to perform poorly in finite samples under weak identification, in particular, the estimator is not guaranteed to have finite moments of any order. We propose the Regularized CUE (RCUE) as a solution to this problem. The RCUE solves a modification of the first-order conditions for the CUE estimator and is shown to be asymptotically equivalent to CUE under many weak moment asymptotics. Our theoretical findings are confirmed by extensive Monte Carlo studies.
| Original language | English (US) |
|---|---|
| Pages (from-to) | 45-57 |
| Number of pages | 13 |
| Journal | Journal of Econometrics |
| Volume | 165 |
| Issue number | 1 |
| DOIs | |
| State | Published - Nov 3 2011 |
Fingerprint
ASJC Scopus subject areas
- Economics and Econometrics
- Applied Mathematics
- History and Philosophy of Science
Cite this
Properties of the CUE estimator and a modification with moments. / Hausman, Jerry; Lewis, Randall; Menzel, Konrad; Newey, Whitney.
In: Journal of Econometrics, Vol. 165, No. 1, 03.11.2011, p. 45-57.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Properties of the CUE estimator and a modification with moments
AU - Hausman, Jerry
AU - Lewis, Randall
AU - Menzel, Konrad
AU - Newey, Whitney
PY - 2011/11/3
Y1 - 2011/11/3
N2 - In this paper, we analyze properties of the Continuous Updating Estimator (CUE) proposed by Hansen et al. (1996), which has been suggested as a solution to the finite sample bias problems of the two-step GMM estimator. We show that the estimator should be expected to perform poorly in finite samples under weak identification, in particular, the estimator is not guaranteed to have finite moments of any order. We propose the Regularized CUE (RCUE) as a solution to this problem. The RCUE solves a modification of the first-order conditions for the CUE estimator and is shown to be asymptotically equivalent to CUE under many weak moment asymptotics. Our theoretical findings are confirmed by extensive Monte Carlo studies.
AB - In this paper, we analyze properties of the Continuous Updating Estimator (CUE) proposed by Hansen et al. (1996), which has been suggested as a solution to the finite sample bias problems of the two-step GMM estimator. We show that the estimator should be expected to perform poorly in finite samples under weak identification, in particular, the estimator is not guaranteed to have finite moments of any order. We propose the Regularized CUE (RCUE) as a solution to this problem. The RCUE solves a modification of the first-order conditions for the CUE estimator and is shown to be asymptotically equivalent to CUE under many weak moment asymptotics. Our theoretical findings are confirmed by extensive Monte Carlo studies.
UR - http://www.scopus.com/inward/record.url?scp=80053314529&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80053314529&partnerID=8YFLogxK
U2 - 10.1016/j.jeconom.2011.05.005
DO - 10.1016/j.jeconom.2011.05.005
M3 - Article
AN - SCOPUS:80053314529
VL - 165
SP - 45
EP - 57
JO - Journal of Econometrics
JF - Journal of Econometrics
SN - 0304-4076
IS - 1
ER -