.png){kind=icon}
返回
SELECTIVITY BIAS CORRECTION METHODS IN POLYCHOTOMOUS SAMPLE SELECTION MODELS
.png){kind=icon}
PRE
AI
DOI：10.1016/0304-4076(94)90039-6.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Polychotomous sample selection models include multinomial choice models and models with multiple rules for sample inclusion. The author analyzes two standard parametric approaches to selectivity bias correction in such models - the Lee and Generalized Heckman (GH) methods. The paper's main point is that Lee's approach, unlike GH, requires strong implicit restrictions on covariances between outcomes and selection indices. A Monte Carlo study demonstrates (1) that the Lee estimator exhibits significant bias when the data-generating process does not conform to its implicit covariance assumptions, and (2) that the GH estimator may have high variance due to multicollinearity between regressors.
Keyword:
SELECTIVITY BIAS
SELF-SELECTION
NONRANDOM SAMPLES
POLYCHOTOMOUS CHOICE
期刊
IF:
4
论文数:
5.3K
被引数:
3.0W
机构
暂无机构信息
引用论文
.png){kind=icon}
分享收藏
查看更多内容