Search Results
Working Paper
Finding Needles in Haystacks: Multiple-Imputation Record Linkage Using Machine Learning
This paper considers the problem of record linkage between a household-level survey and an establishment-level frame in the absence of unique identifiers. Linkage between frames in this setting is challenging because the distribution of employment across establishments is highly skewed. To address these difficulties, this paper develops a probabilistic record linkage methodology that combines machine learning (ML) with multiple imputation (MI). This ML-MI methodology is applied to link survey respondents in the Health and Retirement Study to their workplaces in the Census Business Register. ...
Working Paper
The Impact of Market Factors on Racial Identity: Evidence from Multiracial Survey Respondents
This paper examines the reported race of multiracial persons in the US Current Population Survey (CPS) before 2003, when limited response options exogenously constrained respondents to identify as a single race. Using this survey attribute and the 16-month longitudinal design of the basic monthly CPS, I explore whether market factors help causally determine racial identity. I find that pre-2003 race responds to state-level (1) racial composition, due largely to household composition, and (2) unemployment rates and wages by race. Although these findings suggest potential endogeneity of race, ...
Working Paper
Revisiting the Great Ratios Hypothesis
Kaldor called the constancy of certain ratios stylized facts, whereas Klein and Kosobud called them great ratios. While they often appear in theoretical models, the empirical literature finds little evidence for them, perhaps because the procedures used cannot deal with lack of cointegration, two-way causality and cross-country error dependence. We propose a new system pooled mean group estimator that can deal with these features. Monte Carlo results show it performs well compared with other estimators, and using it on a dataset over 150 years and 17 countries, we find support for five of the ...
Working Paper
Revisiting the Great Ratios Hypothesis
The idea that certain economic variables are roughly constant in the long run is an old one. Kaldor described them as stylized facts, whereas Klein and Kosobud labelled them great ratios. While such ratios are widely adopted in theoretical models in economics as conditions for balanced growth, arbitrage or solvency, the empirical literature has tended to find little evidence for them. We argue that this outcome could be due to episodic failure of cointegration, possible two-way causality between the variables in the ratios and cross-country error dependence due to latent factors. We propose a ...
Discussion Paper
Estimating the output gap in real time
I propose a novel method of estimating the potential level of U.S. GDP in real time. The proposed wage-based measure of economic potential remains virtually unchanged when new data are released. The distance between current and potential output ? the output gap ? satisfies Okun?s law and outperforms many other measures of slack in forecasting inflation. Thus, I provide a robust statistical tool useful for understanding current economic conditions and guiding policymaking.
Working Paper
Aggregation level in stress testing models
We explore the question of optimal aggregation level for stress testing models when the stress test is specified in terms of aggregate macroeconomic variables, but the underlying performance data are available at a loan level. Using standard model performance measures, we ask whether it is better to formulate models at a disaggregated level (?bottom up?) and then aggregate the predictions in order to obtain portfolio loss values or is it better to work directly with aggregated models (?top down?) for portfolio loss forecasts. We study this question for a large portfolio of home equity lines ...
Working Paper
Validating Large Language Model Annotations
This paper proposes a validation framework for LLM-generated measurements when reliable benchmarks are unavailable. Validity is established by testing whether an LLM can reconstruct passages from annotated labels while maintaining semantic consistency with the original text. The framework avoids circular reasoning by establishing testable prerequisite properties that must be met for a validation to be considered successful. Application to news article data demonstrates that the framework serves as a practical alternative to human benchmarking, which offers advantages in objectivity, ...
Working Paper
The Income-Achievement Gap and Adult Outcome Inequality
This paper discusses various methods for assessing group differences in academic achievement using only the ordinal content of achievement test scores. Researchers and policymakers frequently draw conclusions about achievement differences between various populations using methods that rely on the cardinal comparability of test scores. This paper shows that such methods can lead to erroneous conclusions in an important application: measuring changes over time in the achievement gap between youth from high- and low-income households. Commonly-employed, cardinal methods suggest that this ...
Working Paper
In Search of Lost Time Aggregation
In 1960, Working noted that time aggregation of a random walk induces serial correlation in the first difference that is not present in the original series. This important contribution has been overlooked in a recent literature analyzing income and consumption in panel data. I examine Blundell, Pistaferri and Preston (2008) as an important example for which time aggregation has quantitatively large effects. Using new techniques to correct for the problem, I find the estimate for the partial insurance to transitory shocks, originally estimated to be 0.05, increases to 0.24. This larger ...
Working Paper
Estimating (Markov-Switching) VAR Models without Gibbs Sampling: A Sequential Monte Carlo Approach
Vector autoregressions with Markov-switching parameters (MS-VARs) fit the data better than do their constant-parameter predecessors. However, Bayesian inference for MS-VARs with existing algorithms remains challenging. For our first contribution, we show that Sequential Monte Carlo (SMC) estimators accurately estimate Bayesian MS-VAR posteriors. Relative to multi-step, model-specific MCMC routines, SMC has the advantages of generality, parallelizability, and freedom from reliance on particular analytical relationships between prior and likelihood. For our second contribution, we use SMC's ...