Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

Identification of Binary Response Models

Charles F. Manski
Journal of the American Statistical Association
Vol. 83, No. 403 (Sep., 1988), pp. 729-738
DOI: 10.2307/2289298
Stable URL: http://www.jstor.org/stable/2289298
Page Count: 10
  • Download ($14.00)
  • Cite this Item
Identification of Binary Response Models
Preview not available

Abstract

This article studies identification of the threshold-crossing model of binary response. Most research on binary response has considered specific estimators and tests. The study of identification exposes the foundations of binary response analysis by making explicit the assumptions needed to justify different methods. It also clarifies the connections between reduced-form and structural analyses of binary response data. Assume that the binary outcome z is determined by an observable random vector x and by an unobservable scalar u through a model z = 1[ xβ + u ≥ 0]. Also assume that Fu∣ x, the probability distribution of u conditional on x, is continuous and strictly increasing. Given these maintained assumptions, we investigate the identifiability of β given the following restrictions on the distributions (Fu∣ x, x ∈ X): mean independence, quantile independence, index sufficiency, statistical independence, and statistical independence with the distribution known. We find that mean independence has no identifying power. On the other hand, quantile independence implies that β is identified up to scale, provided that the distribution of x has sufficiently rich support. Index sufficiency can identify the slope components of β up to scale and sign, again provided that the distribution of x has a rich support. Statistical independence subsumes both quantile independence and index sufficiency and so implies all of the positive findings previously reported. If u is statistically independent of x with the distribution known, identification requires only that the distribution of x have full rank.

Page Thumbnails

  • Thumbnail: Page 
729
    729
  • Thumbnail: Page 
730
    730
  • Thumbnail: Page 
731
    731
  • Thumbnail: Page 
732
    732
  • Thumbnail: Page 
733
    733
  • Thumbnail: Page 
734
    734
  • Thumbnail: Page 
735
    735
  • Thumbnail: Page 
736
    736
  • Thumbnail: Page 
737
    737
  • Thumbnail: Page 
738
    738