Two-step hierarchical estimation: Beyond regression analysis

被引：63

作者：

Achen, CH ^{[1
]}

机构：

[1] Princeton Univ, Dept Polit, Princeton, NJ 08544 USA

来源：

POLITICAL ANALYSIS | 2005年 / 13卷 / 04期

关键词：

D O I：

10.1093/pan/mpi033

中图分类号：

D0 [政治学、政治理论];

学科分类号：

0302 ; 030201 ;

摘要：

Two-step estimators for hierarchical models can be constructed even when neither stage is a conventional linear regression model. For example, the first stage might consist of probit models, or duration models, or event count models. The second stage might be a nonlinear regression specification. This note sketches some of the considerations that arise in ensuring that two-step estimators are consistent in such cases.

引用

页码：447 / 456

页数：10

共 19 条

[11]

JUSKO KL, 2005, POLITICAL ANAL

[12]

JUSKO KL, 2005, MIDW POL SCI ASS NAT

[13]

KEDAR O, 2005, POLITICAL ANAL

[14]

LEWIS J, 2005, POLITICAL ANAL

[15]

LINDLEY DV, 1972, J ROY STAT SOC B, V34, P1

[16] REGRESSIONS FROM SAMPLES HAVING DIFFERENT CHARACTERISTICS [J].

SAXONHOUSE, GR .

REVIEW OF ECONOMICS AND STATISTICS, 1977, 59 (02) :234-237

[17]

SAXONHOUSE GR, 1976, AM ECON REV, V66, P178

[18]

Snijders T. A. B., 1999, MULTILEVEL ANAL

[19] Causal heterogeneity in comparative research: A Bayesian hierarchical modelling approach [J].

Western, B .

AMERICAN JOURNAL OF POLITICAL SCIENCE, 1998, 42 (04) :1233-1259

← 1 2 →