Search Results
Working Paper
The Dynamic Striated Metropolis-Hastings Sampler for High-Dimensional Models
Having efficient and accurate samplers for simulating the posterior distribution is crucial for Bayesian analysis. We develop a generic posterior simulator called the "dynamic striated Metropolis-Hastings (DSMH)" sampler. Grounded in the Metropolis-Hastings algorithm, it draws its strengths from both the equi-energy sampler and the sequential Monte Carlo sampler by avoiding the weaknesses of the straight Metropolis-Hastings algorithm as well as those of importance sampling. In particular, the DSMH sampler possesses the capacity to cope with incredibly irregular distributions that are full ...
Working Paper
Manufacturing Sentiment: Forecasting Industrial Production with Text Analysis
This paper examines the link between industrial production and the sentiment expressed in natural language survey responses from U.S. manufacturing firms. We compare several natural language processing (NLP) techniques for classifying sentiment, ranging from dictionary-based methods to modern deep learning methods. Using a manually labeled sample as ground truth, we find that deep learning models partially trained on a human-labeled sample of our data outperform other methods for classifying the sentiment of survey responses. Further, we capitalize on the panel nature of the data to train ...
Working Paper
Capturing Macroeconomic Tail Risks with Bayesian Vector Autoregressions
A rapidly growing body of research has examined tail risks in macroeconomic outcomes. Most of this work has focused on the risks of significant declines in GDP, and has relied on quantile regression methods to estimate tail risks. In this paper we examine the ability of Bayesian VARs with stochastic volatility to capture tail risks in macroeconomic forecast distributions and outcomes. We consider both a conventional stochastic volatility specification and a specification featuring a common volatility factor that is a function of past financial conditions. Even though the conditional ...
Working Paper
The Fed's Asymmetric Forecast Errors
I show that the probability that the Board of Governors of the Federal Reserve System staff's forecasts (the "Greenbooks'") overpredicted quarterly real gross domestic product (GDP) growth depends on both the forecast horizon and also whether the forecasted quarter was above or below trend real GDP growth. For forecasted quarters that grew below trend, Greenbooks were much more likely to overpredict real GDP growth, with one-quarter ahead forecasts overpredicting real GDP growth more than 75% of the time, and this rate of overprediction was higher for further ahead forecasts. For forecasted ...
Working Paper
The Inflationary Effects of Sectoral Reallocation
The COVID-19 pandemic has led to an unprecedented shift of consumption from services to goods. We study this demand reallocation in a multi-sector model featuring sticky prices, input-output linkages, and labor reallocation costs. Reallocation costs hamper the increase in the supply of goods, causing inflationary pressures. These pressures are amplified by the fact that goods prices are more flexible than services prices. We estimate the model allowing for demand reallocation, sectoral productivity, and aggregate labor supply shocks. The demand reallocation shock explains a large portion of ...
Working Paper
What's the Story? A New Perspective on the Value of Economic Forecasts
We apply textual analysis tools to measure the degree of optimism versus pessimism of the text that describes Federal Reserve Board forecasts published in the Greenbook. The resulting measure of Greenbook text sentiment, ?Tonality,? is found to be strongly correlated, in the intuitive direction, with the Greenbook point forecast for key economic variables such as unemployment and inflation. We then examine whether Tonality has incremental power for predicting unemployment, GDP growth, and inflation up to four quarters ahead. We find it to have significant and substantive predictive power for ...
Report
A Faster Convergence of Shelter Prices and Market Rent: Implications for Inflation
The Federal Reserve currently faces a “last-mile” problem in bringing inflation back to its 2 percent target. Following the series of federal funds rate hikes that began in March 2022 and ended in July 2023, core (excluding food and energy) Personal Consumption Expenditure (PCE) inflation dropped from a year-over-year peak of 5.6 percent in February 2022 to 2.9 percent in December 2023. At the end of 2023, hopes were high that falling inflation would allow the Fed to cut interest rates several times in 2024. However, the disinflation process slowed noticeably in early 2024, prompting ...
Working Paper
Significance Bands for Local Projections
An impulse response function describes the dynamic evolution of an outcome variable following a stimulus or treatment. A common hypothesis of interest is whether the treatment affects the outcome. We show that this hypothesis is best assessed using significance bands rather than relying on commonly displayed confidence bands. Under the null hypothesis, we show that significance bands are trivial to construct with standard statistical software using the LM principle, and should be reported as a matter of routine when displaying impulse responses graphically.
Newsletter
How Tight is U.S. Monetary Policy
In this Chicago Fed Letter, we use a quantitative macroeconomic model to tackle the question of whether the response of the Federal Reserve (the Fed) to recent high inflation is consistent with its historical behavior. This is an important question because systematic deviations from past behavior could lead the private sector to revise its expectations about how the Fed will respond to inflation going forward, which, according to macroeconomic theory, could affect its ability to stabilize inflation in the future.
Report
A Parsimonious Behavioral SEIR Model of the 2020 COVID Epidemic in the United States and the United Kingdom
I present a behavioral epidemiological model of the evolution of the COVID epidemic in the United States and the United Kingdom over the past 12 months. The model includes the introduction of a new, more contagious variant in the UK in early fall and the US in mid December. The model is behavioral in that activity, and thus transmission, responds endogenously to the daily death rate. I show that with only seasonal variation in the transmission rate and pandemic fatigue modeled as a one-time reduction in the semi-elasticity of the transmission rate to the daily death rate late in the year, the ...