Multi-dimensional Latent Group Structures with Heterogeneous Distributions
Multi-dimensional Latent Group Structures with Heterogeneous Distributions
复制标题
DOI：
10.2139/ssrn.3626938
复制
发表时间：
2020-06
期刊：
影响因子：
--
通讯作者：
X. Leng;Heng Z. Chen;Wendung Wang
中科院分区：
文献类型：
--
作者：
X. Leng;Heng Z. Chen;Wendung Wang
本文旨在识别潜在的分组异质性的分布效果。我们考虑两个面板分位数回归模型与加性固定效应，分位数斜率系数不同组。第一个模型允许分组和时间固定效应，我们提出了一个复合分位数的方法来联合估计组成员，斜率系数和固定效应。我们表明，使用多个分位数提高聚类精度，如果成员是分位数不变的。我们的第二个模型扩展了第一个规格，允许个人和时间固定的影响，我们提出了一个创新的双聚类估计方法。我们应用这些方法来检验管理者激励与风险承担行为之间的关系。
This paper aims to identify the latent grouped heterogeneity of distributional effects. We consider two panel quantile regression models with additive fixed effects, where quantile slope coefficients differ across groups. The first model allows grouped and time fixed effects, and we propose a composite-quantile approach to jointly estimate group memberships, slope coefficients, and fixed effects. We show that using multiple quantiles improves clustering accuracy if memberships are quantile-invariant. Our second model extends the first specification by allowing individual and time fixed effects, and we propose an innovative double-clustering estimation approach. We apply the methods to examine the relationship between managerial incentives and risk-taking behavior.