Characterizing and Testing the Implications of Dynamic Models in Economics and Finance

9409501汉森 本项目开发和实施用于评估动态经济模型的计量经济学方法。 该研究涵盖四个不同的领域。 首先，该项目致力于连续时间随机过程模型的规范和估计。 这类模型在经济学中经常出现，但如何使用离散时间数据来估计和检验它们仍然存在许多悬而未决的问题。 该项目研究了可以回答这些问题中最重要的问题的替代计量经济学方法。 其次，该项目开发了估计线性和对数线性理性预期模型的方法。 这些方法适用的模型包括分配是最优资源分配问题的解决方案的模型以及存在市场或税收扭曲的模型。 研究的估计方法包括半参数程序，从而外生强迫过程的建模灵活的方式。 沿着类似的思路，该项目重新研究了使用滤波方法消除数据低频成分的问题。 这种方法的动机是，动态随机模型的调查抽象的低频运动的时间序列。 所提出的研究的目的是设计的估计方法，一方面，这种错误的规定，但另一方面，有很好的统计特性，是计算上容易处理。 第三，该项目继续致力于开发和调整风险敏感控制理论，以适应动态经济模型。 这种控制理论允许以简单的方式将风险调整纳入模型，以便线性决策规则对经济中的不确定性敏感。 因此，他们抓住了预防性储蓄的动机。 除了测试这些模型的经验，该项目调查的后果，引入异质性的消费者在他们的风险偏好。 最后，该项目研究了英国股票价格和股息的长期历史数据集，并调查了英国股票市场和宏观经济事件之间的联系。

9409501 Hansen This project develops and implements econometric methods for assessing dynamic economic models. The research covers four different areas. First, the project works on specification and estimation of continuous time stochastic process models. Such models occur frequently in economics, but there remains many open questions about how to estimate and test them using discrete time data. The project examines alternative econometric methods that could answer the most important of these questions. Second, the project develops methods for estimating linear and loglinear rational expectations models. The models for which these methods will be applicable include both ones for which the allocations are solutions to optimal resource allocation problems as well as other ones for which there are market or tax distortions. The estimation methods studied include semiparametric procedures whereby the exogenous forcing processes are modeled in a flexible way. Along a similar vein, the project reinvestigates the issue of using filtering methods to eliminate low frequency components of the data. The motivation for such methods is that the dynamic stochastic model under investigation abstracts from the low frequency movements in the time series. The aim of the proposed research is to devise estimation methods that are on the one hand, robust to this misspecification, but on the other hand, have nice statistical properties and are computationally tractable. Third, the project continues work developing and adapting risk-sensitive control theory to models of dynamic economies. This control theory permits risk-adjustments to be incorporated into models in simple ways so that linear decision rules are sensitive to the amount of uncertainty in the economy. Hence they capture a precautionary savings motive. In addition to testing such models empirically, the project investigates the ramifications of introducing heterogeneity among consumers in their risk prefere nces. Finally, the project studies a long historical data set on British stock prices and dividends and investigates the link between the British stock market and macroeconomic events.