### Citations

683 |
Large sample estimation and hypothesis testing,”
- Newey, McFadden
- 1994
(Show Context)
Citation Context ...+ ( sup u∈S0 |r̃(u)− r0(u)| )2] . Assumption 3.2 implies that the density of Ti is positive in the interior of S. Thus, infu∈S0 f0(u) > 0. By uniform consistency of f̃ on S0 (see, e.g., Lemma 8.10 of =-=Newey & McFadden, 1994-=-) the ratio is a OP (1). Thus it suffices to show that supu∈S0 |f̃(u)−f0(u)| = oP (n −1/4) and similarly for r̃. The result follows from Assumption 3.6, the rate condition on hn and Lemma 8.10 of Newe... |

532 |
Self-Selection and Earnings of Immigrants
- Borjas
- 1987
(Show Context)
Citation Context ...ket (Heckman, 1974), union versus nonunion status (Lee, 1978, Robinson & Tomes, 1984), public versus private sector (Dustmann & van Soest, 1998), college attendance (Willis & Rosen, 1979), migration (=-=Borjas, 1987-=-), training program participation (Ashenfelter & Card, 1985, Ham & LaLonde, 1996) as well as occupation (Dolton et al., 1989). The standard Roy model is, however, restrictive in at least two dimension... |

302 | Using the longitudinal structure of earnings to estimate the e¤ect of training programs
- Ashenfelter, Card
- 1985
(Show Context)
Citation Context ...us (Lee, 1978, Robinson & Tomes, 1984), public versus private sector (Dustmann & van Soest, 1998), college attendance (Willis & Rosen, 1979), migration (Borjas, 1987), training program participation (=-=Ashenfelter & Card, 1985-=-, Ham & LaLonde, 1996) as well as occupation (Dolton et al., 1989). The standard Roy model is, however, restrictive in at least two dimensions. First, nonpecuniary aspects matter much in general. For ... |

278 | Some Thoughts on the Distribution of Earnings. Oxford Economic Papers. - Roy - 1951 |

244 | Education and Self Selection”,
- WILLIS, ROSEN
- 1979
(Show Context)
Citation Context ... of participation to the labor market (Heckman, 1974), union versus nonunion status (Lee, 1978, Robinson & Tomes, 1984), public versus private sector (Dustmann & van Soest, 1998), college attendance (=-=Willis & Rosen, 1979-=-), migration (Borjas, 1987), training program participation (Ashenfelter & Card, 1985, Ham & LaLonde, 1996) as well as occupation (Dolton et al., 1989). The standard Roy model is, however, restrictive... |

235 | Semiparametric least squares (SLS) and weighted SLS estimation of single-index models. - Ichimura - 1993 |

225 |
Varieties of selection bias‘,
- Heckman
- 1990
(Show Context)
Citation Context ...rem 2.1 does not provide any result on the location parameters. In general, such parameters are identified only at infinity, i.e. when P (D = 1|X) can be arbitrarily close to zero and one (see, e.g., =-=Heckman, 1990-=-). Theorem 2.1 Suppose that Assumptions 2.1-2.4 hold. Then ψ0 and ψ1 are identified. Alternatively, ψ0 and ψ1 can also be identified at the limit without any exclusion restriction, under the following... |

181 |
Shadow Prices, Market Wages, and Labor Supply’,
- Heckman
- 1974
(Show Context)
Citation Context ...ion, where agents choose the sector which provides them with the higher wage. The idea underlying this model has been very influential in the analysis of choices of participation to the labor market (=-=Heckman, 1974-=-), union versus nonunion status (Lee, 1978, Robinson & Tomes, 1984), public versus private sector (Dustmann & van Soest, 1998), college attendance (Willis & Rosen, 1979), migration (Borjas, 1987), tra... |

178 |
A kernel method of estimating structured nonparametric regression based on marginal integration
- Linton, Nielsen
- 1995
(Show Context)
Citation Context ...res that ψ0(X) (or ψ1(X)) and P (D = 1|X) can vary in a sufficiently independent way. This assumption is weak and is, for instance, assumed implicitly in nonparametric additive regression (see, e.g., =-=Linton & Nielsen, 1995-=-). The first part of Assumption 2.4 covers two rather different situations. The first one is when X0 = X1 = ∅ but we observe some variables which affect the non-pecuniary component but not the potenti... |

169 | An efficient semiparametric estimator of binary response models, Econometrica 61 - Klein, Spady - 1993 |

167 |
Estimating Distributions of Treatment Effects with an Application to the Returns to Schooling and
- Carneiro, Hansen, et al.
- 2003
(Show Context)
Citation Context ...tion value of schooling, which is related to preferences and schooling ability. Recent empirical evidence suggest that these non-pecuniary factors are indeed a key determinant of schooling decisions (=-=Carneiro et al., 2003-=-, and Beffy et al., 2009). Non-pecuniary aspects such as working conditions may also matter when choosing an occupation. Similarly, migration decisions are likely to be driven both by the ex ante mone... |

161 |
The empirical content of the Roy model.
- Heckman, Honore
- 1990
(Show Context)
Citation Context ...erature considering identification of Roy and the closely related competing risks models has produced alternative identification results without such an assumption (see, e.g., Heckman & Honore, 1989, =-=Heckman & Honore, 1990-=- or Abbring & van den Berg, 2003). The identification results we derive in our paper do not rely neither on the aforementioned support condition nor on that independence assumption. Apart from identif... |

148 | Structural equations, treatment effects and econometric policy evaluation”.
- Heckman, Vytlacil
- 2005
(Show Context)
Citation Context ...if no standard instrument is available, we obtain bounds on the distribution of the monetary benefits anticipated by the agents, which correspond in this setting to the marginal treatment effect (see =-=Heckman & Vytlacil, 2005-=-). Standard average treatment effects are point identified if the probability of selection ranges from zero to one, a result in line with the one of Heckman & Vytlacil (2005) in the case of standard i... |

144 |
The Effect of Sample Selection and Initial Conditions in Duration Models: Evidence from Experimental Data on
- Ham, Lalonde
- 1996
(Show Context)
Citation Context ...Tomes, 1984), public versus private sector (Dustmann & van Soest, 1998), college attendance (Willis & Rosen, 1979), migration (Borjas, 1987), training program participation (Ashenfelter & Card, 1985, =-=Ham & LaLonde, 1996-=-) as well as occupation (Dolton et al., 1989). The standard Roy model is, however, restrictive in at least two dimensions. First, nonpecuniary aspects matter much in general. For instance, in the cont... |

135 | Dynamic discrete choice and dynamic treatment effects,” - Heckman, Navarro - 2007 |

85 | Inference on Regressions with Interval Data on a Regressor or Outcome,” - Manski, Tamer - 2002 |

84 | Two-Step Series Estimation of Sample Selection Models,” Working Paper, - Newey - 1988 |

65 | Social interactions in high school: lessons from an earthquake, - Cipollone, Rosolia - 2007 |

64 |
The identifiability of the competing risks model
- HECKMAN, HONORÉ
- 1989
(Show Context)
Citation Context ...ive, and much of the literature considering identification of Roy and the closely related competing risks models has produced alternative identification results without such an assumption (see, e.g., =-=Heckman & Honore, 1989-=-, Heckman & Honore, 1990 or Abbring & van den Berg, 2003). The identification results we derive in our paper do not rely neither on the aforementioned support condition nor on that independence assump... |

42 | Public and private sector wages of male workers in Germany, - Dustmann, Soest - 1998 |

39 | Identifying and Estimating the Distributions of Ex Post and Ex Ante Returns to Schooling: A Survey of Recent Developments", unpublished manuscript
- Cunha, Heckman
- 2006
(Show Context)
Citation Context ...by the ex ante monetary returns and the psychic costs associated with the decision to migrate (Bayer et al., 2008). Second, as emphasized by a recent stream of the literature on schooling choice (see =-=Cunha & Heckman, 2007-=-, for a survey), agents most often do not anticipate perfectly their potential earnings in each sector at the moment of their decision. Because of ex ante uncertainty, their decision depends on expect... |

29 |
Union Wage Differentials in the Public and Private Sectors: A Simultaneous Equations Specification’.
- Robinson, Tomes
- 1984
(Show Context)
Citation Context ...ith the higher wage. The idea underlying this model has been very influential in the analysis of choices of participation to the labor market (Heckman, 1974), union versus nonunion status (Lee, 1978, =-=Robinson & Tomes, 1984-=-), public versus private sector (Dustmann & van Soest, 1998), college attendance (Willis & Rosen, 1979), migration (Borjas, 1987), training program participation (Ashenfelter & Card, 1985, Ham & LaLon... |

25 |
Bounds for a joint distribution function with fixed subdistribution functions: application to competing risks,”
- Peterson
- 1976
(Show Context)
Citation Context ...ly worthwile since it conveys information about the relative importance of general vs. specific human capital. This dependence, which has received much attention in competing risks models (see, e.g., =-=Peterson, 1976-=-, van den Berg, 1997, Abbring & van den Berg, 2003), has been identified in generalized Roy models by imposing a factor model (see Carneiro et al., 2003). However, it would be interesting to conduct a... |

23 | Occupational Choice and Earnings Determination: The Role of Sample Selection and Non-Pecuniary Factors. - Dolton, Makepeace, et al. - 1989 |

22 | Social interactions and schooling decisions, - Lalive, Cattaneo - 2009 |

21 | Identification of Treatment Effects Using Control Functions in Models with Continuous, Endogenous Treatment and Heterogeneous Effects, Econometrica, forthcoming - Florens, Heckman, et al. - 2008 |

15 |
Choosing the Field of Study in Postsecondary Education
- Beffy, Fougre, et al.
(Show Context)
Citation Context ...ich is related to preferences and schooling ability. Recent empirical evidence suggest that these non-pecuniary factors are indeed a key determinant of schooling decisions (Carneiro et al., 2003, and =-=Beffy et al., 2009-=-). Non-pecuniary aspects such as working conditions may also matter when choosing an occupation. Similarly, migration decisions are likely to be driven both by the ex ante monetary returns and the psy... |

15 | 2009), “Estimating Distributions of Potential Outcomes using Local Instrumental Variables with an Application to Changes in College Enrolment and Wage Inequality - Carneiro, Lee |

13 |
Training and lifetime income
- Kuruscu
- 2006
(Show Context)
Citation Context ...be independently and identically distributed over time, with mean zero. We introduce the dummy 1{t ≤ B} to account for non significant marginal returns to experience after B years of work (see, e.g., =-=Kuruscu, 2006-=-, for similar assumption on wage growth). We also suppose that νk,t is independent of D, so that ρk is simply identified by ρk = E(yk,t− yk,t−1|D = k), for t ≤ B. Then, we can compute when D = k the f... |

12 | Identi of a competing risks model with unknown transformations of latent failure times
- Lee
- 2006
(Show Context)
Citation Context ...ng it to ηk does not modify the model. Assumption 2.3 is a technical condition which is usual in competing risks or Roy models (see in particular Heckman & Honore, 1990, for the case of Roy model and =-=Lee, 2006-=- for the case of competing risks models.). 2.2 Identification of (ψ0, ψ1) Before detailing our key result on the identification of G, we present in this subsection two strategies to recover (ψ0, ψ1). ... |

10 | 2008): “Nonparametric Identification and Estimation in a Generalized Roy Model,” NBER Working Paper Series
- Bayer, Khan, et al.
(Show Context)
Citation Context ...may also matter when choosing an occupation. Similarly, migration decisions are likely to be driven both by the ex ante monetary returns and the psychic costs associated with the decision to migrate (=-=Bayer et al., 2008-=-). Second, as emphasized by a recent stream of the literature on schooling choice (see Cunha & Heckman, 2007, for a survey), agents most often do not anticipate perfectly their potential earnings in e... |

10 | Estimation of the Binary Response Model Using a Mixture of Distributions Estimator (MOD
- Coppejans
- 2001
(Show Context)
Citation Context ...considered in Subsection 5.5. We estimate the model relying on the three-stages semiparametric procedure detailed in Section 3. More precisely, we use for the first step a mixture of probit (see e.g. =-=Coppejans, 2001-=-) with K1 = 3 mixture components.22 The second step is performed with Newey (2009)’s series estimator, with K2 = 9 approximating terms. δ0 is finally estimated with the same specifications as in the M... |

6 | Association measures for durations in bivariate hazard rate models - Berg - 1997 |

5 | A new instrumental method for dealing with endogenous selection - D’Haultfoeuille - 2010 |

5 | Another look at the identification at infinity of sample selection models - d’Haultfoeuille, Maurel - 2009 |

5 |
Unionism and relative wage rates: A simultaneous equations model with qualitative and limited dependent variables
- Lee
- 1978
(Show Context)
Citation Context ...ides them with the higher wage. The idea underlying this model has been very influential in the analysis of choices of participation to the labor market (Heckman, 1974), union versus nonunion status (=-=Lee, 1978-=-, Robinson & Tomes, 1984), public versus private sector (Dustmann & van Soest, 1998), college attendance (Willis & Rosen, 1979), migration (Borjas, 1987), training program participation (Ashenfelter &... |

5 | Identification in accelerated failure time competing risks models. Working Paper - Lee, Lewbel - 2009 |

4 | The identifiability of the mixed proportional hzards competing risks model - Abbring, Berg - 2003 |