Business Fixed Investment - Optimal Instrumental Variables Estimation - Tests of Predictive Ability

企业固定投资 - 最佳工具变量估计 - 预测能力测试

• 批准号: 9709756
• 负责人: Kenneth West
• 金额: $ 21.91万
• 依托单位: University of Wisconsin-Madison
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1997
• 资助国家: 美国
• 起止时间: 1997-07-15 至 2001-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9709756&HistoricalAwards=false
• 关键词: Business; Fixed; Investment; Optimal; Instrumental

This project consists of two part which are of roughly equal magnitude. The first part involves empirical research on business fixed investment. The aim here is to estimate and interpret dynamic models of investment. Models and solution techniques will be developed for analyzing the response of investment to movements in interest rates, wages, taxes, and productivity. The model for investment will be embedded in a larger macroeconomic model, to study the economy-wide links between investment on the one hand and business cycles and economics policy on the other. The second part of the project involves theoretical econometrics, with substantive empirical applications. The main effort here concerns optimal instrumental variables estimation in the presence of conditionally heteroskedastic disturbances. The author will develop and evaluate, via simulations, feasible estimators, applying the estimators to weekly data on exchange and interest rates. A related but smaller project will continue ongoing work on inference about out-of-sample predictions, by developing regression based tests and allowing for a broader class of measures of predictive ability.

这个项目由两部分组成，两部分的规模大致相当。 第一部分是企业固定资产投资的实证研究。 这里的目的是估计和解释投资的动态模型。 将开发模型和解决方案技术，用于分析投资对利率，工资，税收和生产率变动的反应。 投资模型将嵌入一个更大的宏观经济模型，以研究投资与商业周期和经济政策之间的整体经济联系。 该项目的第二部分涉及理论计量经济学，具有实质性的实证应用。 这里的主要工作是在存在条件异方差干扰的情况下，最优工具变量估计。 作者将开发和评估，通过模拟，可行的估计，应用估计每周数据的汇率和利率。 一个相关但规模较小的项目将继续进行关于样本外预测的推理工作，方法是开发基于回归的测试，并允许更广泛的预测能力测量。