Estimating Earnings Dynamics with Micro Data

这项研究创建了日本男性收入的面板数据集，这些数据来自家庭经济研究所编制的日本消费者小组调查，然后试图估计收入动态模型。在考虑了样本选择程序、测量误差、教育程度和职业差异等因素后，我们发现，个人资料异质性模型不能在统计学上被拒绝。这一发现与最近美国基于数据的辩论一致，即现有的高度持续盈利冲击的假设可能是不合适的。

This study creates panel data sets of Japanese male earnings from the Japanese Panel Survey of Consumers compiled by the Institute for Research on Household Economics, and then attempts to estimate earnings dynamics models. After taking into account various factors such as sample selection procedures, measurement error, educational backgrounds, and occupational differences, we find that the profile heterogeneity model cannot be statistically rejected. This finding is consistent with the recent US data-based debate that the existing assumption of highly persistent earnings shocks may be inappropriate.