Asset Returns and Economic Fluctuations

一般来说，人们很自然地认为，经济生产过程中的风险和回报应随着经济波动的过程而变化，而资产回报应反映这种变化。然而，资产回报与经济波动相联系的机制尚未得到很好的理解。本项目的贡献在于开发了经验上成功的资产回报与经济波动之间联系的经济模型。这些模型之所以重要，是因为：1)它们将有助于区分不同的商业周期理论；2)它们是严格确定金融市场波动是否表明投机趋势或有效市场对基本面变化的反应的唯一方法之一；3)成功的经济模型可以改善投资组合配置规则。该项目的第一部分扩展了一个基于生产的资产定价模型，这是一个经济中生产过程的物理回报率模型。该扩展涵盖了使用几种不同的技术和资产价格的横截面数据对许多不同资产的收益同时变化的实证检验；技术描述的改进；以及对时间聚合的明确研究。该项目的第二部分涉及计算资产收益条件矩变化所隐含的贴现因子条件矩的变化。这些结果被用来确定资产回报的行为是由于“潮流”还是由于贴现因素的理性运动。

At a general level, it is natural to expect that risks and rewards in the economy's production processes should vary over the course of economic fluctuations, and that asset returns should mirror this variation. However, the mechanism by which asset returns are linked to economic fluctuations is not yet well understood. The contribution of this project comes from developing empirically successful economic models of the links between asset returns and economic fluctuations. Such models are important because: 1) they will help discriminate between competing theories of business cycles; 2) they are one of the only ways of rigorously determining whether or not volatility in financial markets indicate speculative fads or the response of efficient markets to changes in fundamentals; and 3) successful economic models could improve portfolio allocation rules. The first part of this project extends a production-based asset pricing model, that is a model of the physical rates of return of production processes in the economy. The extension covers empirical tests of simultaneous variations in the returns on many different assets using several different techniques and cross- sectional data of asset prices; improvements in the description of technology; and the explicit study of time aggregation. The second part of the project involves the calculation of the variation in conditional moments of discount factors implied by variation in the conditional moments of asset returns. The results are used to determine whether or not the behavior of asset returns is due to "fads" or to rational movements in discount factors.