A copula with a flexibly dependence structure can capture complexity and heterogeneity in economic and financial time series. Based on the recently proposed single-index copula, we propose a simultaneous variable selection and estimation procedure. This method allows for choosing the most relevant state variables by using a penalized estimation with large sample properties derived. Simulation results demonstrate the good performance of the method in selecting relevant state variables and estimating unknown index coefficients and dependence parameters. We apply the proposed procedure to four states' housing markets in the United States and identify six macroeconomic factors that drive their dependence structure.

中文翻译：

---

单指数copula模型的半参数估计和变量选择

具有灵活依赖结构的 copula 可以捕捉经济和金融时间序列中的复杂性和异质性。基于最近提出的单指数 copula，我们提出了一个同步变量选择和估计程序。该方法允许通过使用具有大样本属性的惩罚估计来选择最相关的状态变量。仿真结果表明，该方法在选择相关状态变量和估计未知指标系数和相关参数方面具有良好的性能。我们将建议的程序应用于美国四个州的房地产市场，并确定了推动其依赖结构的六个宏观经济因素。