发帖
雷达卡
摘要翻译：
通常使用有序的probit和logit模型来估计组间幸福结果的平均等级。然而，最近强调的是，在大多数应用程序中没有识别这种排名。然后，我们能从标准统计软件如STATA报告的群体间的平均排名中学到任何东西吗？我们认为它可以被解释为中位秩，即使当平均秩不是时，也可以识别中位秩。因此，我们建议将注意力集中在按中位数而不是平均值对幸福结果（和其他序数数据）进行排名上。中值排序也可以半参数化进行，并给出了一种新的约束混合整数优化方法。为了说明这一点，我们使用一般的社会调查数据来展示在1972年至2006年间幸福文学中对美国所持的著名的伊斯特林悖论。
---
英文标题：
《Robust Ranking of Happiness Outcomes: A Median Regression Perspective》
---
作者：
Le-Yu Chen, Ekaterina Oparina, Nattavudh Powdthavee, Sorawoot Srisuma
---
最新提交年份：
2021
---
分类信息：
一级分类：Economics 经济学
二级分类：Econometrics 计量经济学
分类描述：Econometric Theory, Micro-Econometrics, Macro-Econometrics, Empirical Content of Economic Relations discovered via New Methods, Methodological Aspects of the Application of Statistical Inference to Economic Data.
计量经济学理论，微观计量经济学，宏观计量经济学，通过新方法发现的经济关系的实证内容，统计推论应用于经济数据的方法论方面。
--
---
英文摘要：
The mean rank of happiness outcomes between groups has often been estimated using ordered probit and logit models. However, it has recently been highlighted that such ranking is not identified in most applications. Can we then learn anything from a mean rank between groups that is reported by standard statistical softwares such as STATA? We argue it can instead be interpreted as the median rank, which is identified even when the mean rank is not. We thus suggest focusing on ranking happiness outcomes (and other ordinal data) by the median rather than the mean. The median ranking can also be performed semiparametrically and we provide a new constrained mixed integer optimization procedure for implementation. To illustrate, we use General Social Survey data to show the well-known Easterlin Paradox in the happiness literature holds for the US over the period 1972 to 2006.
---
PDF链接：
https://arxiv.org/pdf/1902.07696
京公网安备 11010802022788号
