Advances in Macroeconomics and Econometrics

这项研究产生了新的见解，企业的价格制定行为，并制定了指导方针，将有助于在广泛的应用计量经济学工作。提出了两个研究领域。 第一个项目研究的问题是，为什么企业不立即调整价格以适应新的经济基本面？这个问题触及了经济学中许多核心问题的核心。这里的新建议是研究一个长时间序列，其中有足够数量的观察到的价格变化，我们可以对预测单个公司改变价格的条件做出有用的统计陈述。一个基本情况下的菜单成本模型的含义，过去的历史价格和基本面应该影响观察价格变化的概率，只有通过当前价格和基本面之间的差距差距。提出了一个框架来检验这一含义，并量化和解释在公司的实际price.The第二组项目中观察到的替代dynamics研究的作用，在计量经济学的正常化。这个问题，长期以来被认为是无关的实质性问题，最近已被证明是一个显着的差异，估计的脉冲响应函数和置信区间，通常计算在一些模型。这项研究建议提出了一个统一的原则，可以用来解决已经记录的问题，以及在一些计量经济学的应用程序中，这个问题还没有得到承认的帮助。

This research generates new insights into firms' price-setting behavior and develops guidelines that will be helpful in a wide range of applied econometric work. Two areas of research are proposed. The first project studies the question, why don't firms instantly adjust their prices to new economic fundamentals? This question gets to the heart of many of the central issues in economics. The novel suggestion here is to study a long time series with a sufficient number of observed price changes that we can make a useful statistical statement about the conditions under which an individual firm is predicted to change its price. A base-case menu cost model is shown to have the implication that the past history of prices and fundamentals should influence the probability of observing a price change only through the gap between the current price and fundamental value. A framework is proposed for testing this implication and quantifying and interpreting alternative dynamics that are observed in firms' actual prices.The second set of projects studies the role of normalization in econometrics. This issue, long thought to be irrelevant for substantive questions, has recently been shown to make a significant difference for the estimated impulse-response functions and confidence intervals as typically calculated in a number of models. This research proposal suggests a unifying principle that could be used to resolve both the problems that have already been documented as well as help in a number of econometric applications where this issue has yet to be recognized.