I joined the faculty of the Institute of Economics, Academia Sinica in 2018 as an assistant research fellow after receiving my PhD in Economics from UC Berkeley. My research interest is in Econometrics.
The maximum utility estimation proposed by Elliott and Lieli (2013) can be viewed as cost-sensitive binary classification; thus, its in-sample overfitting issue is similar to that of perceptron learning. A utility-maximizing prediction rule (UMPR) is constructed to alleviate the in-sample overfitting of the maximum utility estimation. We establish non-asymptotic upper bounds on the difference between the maximal expected utility and the generalized expected utility of the UMPR. Simulation results show that the UMPR with an appropriate data-dependent penalty achieves larger generalized expected utility than common estimators in the binary classification if the conditional probability of the binary outcome is misspecified.
This paper proposes a nonparametric estimator of the counterfactual copula of two outcome variables that would be affected by a policy intervention. The proposed estimator allows policymakers to conduct ex-ante evaluations by comparing the estimated counterfactual and actual copulas as well as their corresponding measures of association. Asymptotic properties of the counterfactual copula estimator are established under regularity conditions. These conditions are also used to validate the nonparametric bootstrap for inference on counterfactual quantities. Simulation results indicate that our estimation and inference procedures perform well in moderately sized samples. Applying the proposed method to studying the effects of college education on intergenerational income mobility under two counterfactual scenarios, we find that while providing some college education to all children is unlikely to promote mobility, offering a college degree to children from less educated families can significantly reduce income persistence across generations.
"Utility-Maximizing Binary Prediction via the Nearest Neighbor Method and Its Application to Credit Scoring"
We propose nonparametric k-nearest neighbor prediction rules under the framework of utility-maximizing binary prediction with possibly many predictors. One of these prediction rules, with an attempt to ‘break’ the curse of dimensionality, is constructed based on the predictors selected by variable selection methods. We establish that these prediction rules, allowing for the data-dependent selection of parameter k, are utility consistent under regularity assumptions. Such utility consistency is confirmed by the simulation results. We illustrate these prediction rules with an application to credit scoring in peer-to-peer lending and find that common predictors of the business cycle yield limited improvement in profitability.
The forecast combination puzzle is often found in literature: The equal-weight scheme tends to outperform sophisticated methods of combining individual forecasts. Exploiting this finding, we propose a hedge egalitarian committees algorithm (HECA), which can be implemented via mixed integer quadratic programming. Specifically, egalitarian committees are formed by the ridge regression with shrinkage toward equal weights; subsequently, the forecasts provided by these committees are averaged by the hedge algorithm. We establish the no-regret property of HECA. Using data collected from the ECB Survey of Professional Forecasters, we find the superiority of HECA relative to the equal-weight scheme during the COVID-19 recession.
We propose a counterfactual Kaplan-Meier estimator that incorporates exogenous covariates and unobserved heterogeneity of unrestricted dimensionality in duration models with random censoring. Under some regularity conditions, we establish the joint weak convergence of the proposed counterfactual estimator and the unconditional Kaplan-Meier (1958) estimator. Applying the functional delta method, we make inference on the cumulative hazard policy effect, that is, the change of duration dependence in response to a counterfactual policy. We also evaluate the finite sample performance of the proposed counterfactual estimation method in a Monte Carlo study.
"Testing Monotonicity in a Model with Nonseparable Time-Invariant Heterogeneity"
This paper develops a test for structural monotonicity, that is, monotonicity of a structural function in an explanatory variable given any observable covariates and nonseparable time-invariant unobserved heterogeneity. We show that in a two-period panel data model, under some conditions, structural monotonicity implies shape constraints on the joint cumulative distribution function (CDF) of outcome variables conditional on the explanatory variables and covariates over specific regions. These regions are parameterized by a nuisance parameter, which can be consistently estimated. We propose a test for structural monotonicity according to the shape constraints on the conditional joint CDF over the estimated regions, and validate the empirical bootstrap method under some high-level conditions. Some Monte Carlo experiments show that the proposed test can detect departures from structural monotonicity, which are not revealed by some tests for regression monotonicity, for example tests proposed by Ghosal, Sen, and van der Vaart (2000) and Chetverikov (2017).
"A Multivariate Distribution with Pareto Tails and Pareto Maxima" (with Costas Arkolakis and Andrés Rodríguez-Clare)
We present a new multivariate distribution with Pareto distributed tails and maxima. The distribution has a number of properties that make it useful for applied work. Compared to uncorrelated univariate Pareto distributions, the distribution features one additional parameter that governs the covariance of its realizations. We show that this distribution is indeed valid by proving a general result about $n$-increasing functions.