2010
DOI: 10.2139/ssrn.1625890
|View full text |Cite
|
Sign up to set email alerts
|

Inference Based on Conditional Moment Inequalities

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

5
278
0
2

Year Published

2010
2010
2023
2023

Publication Types

Select...
5
2

Relationship

1
6

Authors

Journals

citations
Cited by 138 publications
(285 citation statements)
references
References 35 publications
5
278
0
2
Order By: Relevance
“…The model (4.11) appears almost the same as the conditional moment inequality model discussed extensively in, e.g., Andrews and Shi (2013) and Chernozhukov, Lee, and Rosen (2008). However there is one essential difference that we need now address.…”
Section: Aggregating Moment Inequalities To the Market Levelmentioning
confidence: 91%
See 1 more Smart Citation
“…The model (4.11) appears almost the same as the conditional moment inequality model discussed extensively in, e.g., Andrews and Shi (2013) and Chernozhukov, Lee, and Rosen (2008). However there is one essential difference that we need now address.…”
Section: Aggregating Moment Inequalities To the Market Levelmentioning
confidence: 91%
“…This procedure makes it feasible in practice to perform inference in moment inequality models with many parameters and is critically needed for our demand estimation problem. The existing approach to inference in a moment inequality setting such as ours -for example, Andrews and Shi (2013) -requires exhaustive grid search over the parameter space to compute confidence sets. However, such computation is infeasible for demand studies because at least a moderate number of control variables are needed to ensure validity of the instrument for price, resulting in a moderate to large dimensional parameter space.…”
Section: Introductionmentioning
confidence: 99%
“…However, we show that this computational hardship can be avoided, and the sharp identi…cation region can be characterized as the set of parameter values for which the maximum of an easy-to-compute superlinear (hence concave) function over the unit ball is equal to zero. We show that …nite sample inference can be carried out adopting the generalized moment selection procedure with in…nitely many conditional moment inequalities recently proposed by Andrews and Shi (2009). We exemplify our methodology by applying it to empirically relevant models, for which a feasible characterization of I was absent in the literature.…”
Section: Discussionmentioning
confidence: 94%
“…Remark 2.1 Using the method proposed by Andrews and Shi (2009), expression (2.4) can be transformed, using appropriate instruments, into a set of unconditional moment inequalities indexed by the instruments and by u 2 B; even when the conditioning variables have a continuous distribution. Equations (2.2)-(2.3) can be modi…ed accordingly, to yield straightforward criterion functions which are minimized by every parameter in the sharp identi…cation region.…”
Section: Representation Through Random Sets Theorymentioning
confidence: 99%
See 1 more Smart Citation