Job Market Candidates

Professor Shurojit CHATTERJI
Email: shurojitc@smu.edu.sg
Tel: +65 68280314

Professor Jingyi XUE
Email:  jyxue@smu.edu.sg
Tel: +65 68085466

Email: tkunimoto@smu.edu.sg
Tel: +65 68289558

WORKING PAPERS

"Compellingness in Nash Implementation",  with Takashi Kunimoto and Shurojit Chatterji (Job Market Paper)

A social choice function (SCF) is said to be Nash implementable if there exists a mechanism in which every Nash equilibrium induces outcomes specified by the SCF. The main objective of this paper is to assess the impact of considering mixed strategy equilibria in Nash implementation. We call a mixed strategy equilibrium “uncompelling” if its outcome is strictly worse for any agent than that induced by the SCF. We show that if the finite environment and the SCF to be implemented jointly satisfy what we call Condition COM, we construct a finite mechanism which Nash implements the SCF in pure strategies and its any mixed strategy Nash equilibrium outcome is either uncompelling or consistent with the SCF. Our mechanism has several desirable features: transfers can be completely dispensable; only finite mechanisms are considered; integer games are not invoked; and agents’ attitudes toward risk do not matter. These features make our result quite distinct from many other prior attempts to handle mixed strategy equilibria in the theory of implementation.

RESEARCH PAPERS

"Minimum Reversals Domains: a link between median rules and monotonicity”, (revised and resubmitted at Games and Economic Behavior)

We show that any monotonic, anonymous, unanimous and tops-only rule defined on a MAT-connected domain must be a median voter rule on a tree. Moreover, a median voter rule on a tree is monotonic if and only if the domain is a minimum reversals domain, which is a weakening of known preference domains. These two results are combined to obtain a set of necessary and sufficient conditions for the existence of monotonic, anonymous, and unanimous rules on a MAT-connected domain.

Professor Yichong ZHANG
Email: yczhang@smu.edu.sg
Tel: +65 68280881

Professor Jun YU
Email: yujun@smu.edu.sg
Tel: +65 68280858

Email: jiali@smu.edu.sg
Tel: +65 68280890

Professor Wenjie WANG
Email: wang.wj@ntu.edu.sg
Tel: +65 6316-8958

WORKING PAPERS

"A Valid Anderson-Rubin Test under Both Fixed and Diverging Number of Weak Instruments",  (Job Market Paper)

The conventional and jackknife Anderson-Rubin (AR) Tests are developed separately to conduct weak-identification-robust inference when the number of instrumental variables (IVs) are fixed or diverging to infinity with the sample size, respectively. These two tests compare distinct test statistics with distinct critical values. To implement them, researchers first need to take a stance on the asymptotic behaviour of the number of IVs, which is ambiguous when this number is just moderate. Instead, in this paper, we propose two analytical weak-identificationrobust  AR tests, both of which control asymptotic size whether the number of IVs are fixed or diverging. We further analyze the power properties of these uniformly valid AR tests under both cases.

RESEARCH PAPERS

"A Conditional Linear Combination Test with Many Weak Instruments"

We consider a linear combination of jackknife Anderson-Rubin (AR), jackknife Lagrangian multiplier (LM), and orthogonalized jackknife LM tests for inference in IV regressions with many weak instruments and heteroskedasticity. Following I.Andrews (2016), we choose the weights in the linear combination based on a decision-theoretic rule that is adaptive to the identification strength. Under both weak and strong identifications, the proposed test controls asymptotic size and is admissible among certain class of tests. Under strong identification, our linear combination test has optimal power against local alternatives among the class of invariant or unbiased tests which are constructed based on jackknife AR and LM tests. Simulations and an empirical application to Angrist and Krueger’s (1991) dataset confirm the good power properties of our test.

PUBLICATIONS

"A Conditional Linear Combination Test with Many Weak Instruments", with Zhang Yichong and Wang Wenjie, Journal of Econometrics, forthcoming

WORKING PAPERS

"Uniform Inference for High-Frequency Data",  (Job Market Paper)

We address the uniform inference problem for high-frequency data that includes prices, volumes, and trading ows. Such data is modeled with a general state-space framework, where latent state process is the corresponding risk indicators, e.g., volatility, price jump, average order size, and arrival of events. The functional estimators are formed as the collection of localized estimates across different time points. Although the proposed estimators do not admit a functional central limit theorem, a Gaussian strong approximation, or coupling, is established under in-fill asymptotics to facilitate feasible inference. We apply the proposed methodology to distinguish the informative part from the Federal Open Market Committee speeches, and to analyze the impact of social media activities on cryptocurrency markets.

RESEARCH PAPERS

“Optimal Nonparametric Range-Based Volatility Estimation”, with Tim Bollerslev and Jia Li, accepted in Journal of Econometrics.

We present a general framework for optimal nonparametric spot volatility estimation based on intraday range data, comprised of the first, highest, lowest, and last price over a given time interval. We rely on a decision-theoretic approach together with a coupling-type argument to directly tailor the form of the nonparametric estimator to the specific volatility measure of interest and relevant loss function. The resulting new optimal estimators offer substantial efficiency gains compared to existing commonly used range-based procedures.

PUBLICATIONS

“Permutation-based Tests for Discontinuities in Event Studies”, with Federico Bugni and Jia Li, Quantitative Economics, 14(1), 2023, 37-70.

“Seemingly Unrelated Regression Estimation for VAR Models with Explosive Roots”, with Ye Chen and Jian Li, Oxford Bulletin of Economics and Statistics, 85(1), 2023, 910-937.

Professor Pao-Li CHANG
Email: plchang@smu.edu.sg
Tel: +65 68280830

Professor Jianhuan XU
Email: jhxu@smu.edu.sg
Tel: +65 68280085

Email: linma@smu.edu.sg
Tel: +65 68280876

WORKING PAPERS

"US Foreign Trade Zones - Working as Cushions for Trade War",  (Job Market Paper)

Enabling deferral or elimination of duty payments, the US Foreign Trade Zones (FTZs) displayed significant “Cushion Effects” for producers within the zones during the US-China trade war. The first source of “Cushion Effects” resulted from the over 28% increase in the zones’ export volume during the tariff war, measured by the extra duties directly exempted. The effect amounted to about 883 million dollars in 2019. In addition, the FTZ demand for sanctioned components used in the production of domestically sold products was less affected due to the deferral and efficiency of duty payments, providing the second source of “Cushion Effects”. I apply the rarely quantitatively analyzed FTZ import data from USITC and compiled trade volumes from zones’ annual reports. The empirical identification results at the intensive margin show that tariff shocks triggered more sales of FTZ firms to both foreign and domestic markets. This phenomenon is especially pronounced among the 120 new firms whose entrance was positively correlated with extra tariffs, providing the evidence of the existence of “Cushion Effects” at the extensive margin. The supplementary duties that were exempted, temporarily deferred, and non-paid by the year’s end quantify “Cushion Effects”. Under the protection, FTZ firms’ tendency to pre-storage when anticipating new tariffs or substitute the domestic and non-affected foreign sources of inputs for their sanctioned Chinese counterparts is less pronounced, as the FD and DID models estimate. For the cutting-edge technology inputs of List 2 issued in Section 301 Act, which were also included in the “Made in China 2025” program, the imposed tariff shocks generated positive impacts on FTZ producers’ import volumes. Lastly, the empirical observations are mapped theoretically to a two-tier Melitz-type model, and the counterfactual comparative statistics derived provide a constructive suggestion that the government can enhance the protection by relaxing the criteria of entry into the zones.

RESEARCH PAPERS

"Determinants of FDI Entry into US Foreign Trade Zones"

Up to 2021, a total of 442 production firms have been established within the US Foreign Trade Zones (FTZs), encompassing a diverse spectrum of 144 NAICS6 sectors, predominantly affiliated with the manufacturing industry. By examining the ownership structure of FTZ producers at the headquarters level, I find that approximately 26% of them are subsidiaries of foreign parent companies, with Japan representing the largest source of FDI among these entities. To account for the inherent heterogeneity of foreign investments across industries within the zones, this study presents a comprehensive model that encompasses variations in headquarter service intensity, foreign component intensity, and productivity. By employing the compiled dataset, empirical verification is conducted to validate the propositions implied by the model. The findings demonstrate that non-US headquarters exhibit a stronger propensity to enter FTZs when operating within sectors characterized by intensive usage of dutiable inputs. In contrast, local producers display stronger responses to sectoral productivity enhancements.

“Uncertain Programming Model for Shortest Path Problem”, with Soleimani-Alyar M.

The shortest path problem is concerned with finding a path from the source node to the sink node with a minimum total length. It is however a difficult task to find the shortest path in an indeterministic network, especially when there are insufficient samples or historical data. This paper considers the shortest path problem with uncertain lengths and introduces a new uncertain programming model for it. This model can find the ideal shortest path on an uncertain network and is suitable for cases where there is no access to expert estimates of the value of belief degree. To model this problem, we use both uncertainty theory and Dijkstra’s algorithm, to obtain the best distribution function of minimum lengths by empirical distribution function.

“The Determination of Agricultural Insurance Rate Based on Satellite Image and Yield Simulation Spatial Model”

The main content of this paper is to combine the unit area yield estimation model of satellite image processing and the yield simulation spatial model based on the Kriging interpolation method. The rate of agricultural production insurance in the whole research area in 2018 is determined. Because the historical reimbursement data of domestic agricultural insurance are scarce, we adopted and improved upon the traditional yield simulation model used in the premium rate determination. I propose methods to address the shortcomings of the traditional model.

PUBLICATIONS

"Intelligent Growth Model of City", with K. Huang, C. Liang. Growth, 8, 2017, 382-383

Professor Yuan MEI
Email: yuanmei@smu.edu.sg
Tel: +65 68085212

Professor Jing LI
Email: lijing@smu.edu.sg
Tel: +65 68085454

Email: linma@smu.edu.sg
Tel: +65 68280876

WORKING PAPERS

"Educational Migration in China",  with Lin Ma (Job Market Paper)

Educational resources are distributed unevenly across space and could contribute to spatial inequality. We develop a dynamic spatial model with life-cycle elements to study the impacts of location-specific educational resources. In the model, individuals determine whether and where to attend college, weighing on the distance to home, the expected option value of education, and the educational resources in the destination. Locations with more colleges attract more students. Moreover, as mobility costs increase with age, many college graduates stay in the city of their alma mater, leading to long-term changes in skill composition. We quantify the model to the context of China and structurally estimate the cost of obtaining a college degree in each location. We show that the college expansion between 2005 and 2015 had minimal impacts on welfare and skill composition, as it diverts resources towards the locations already well-endowed with colleges. More evenly distributed colleges could improve aggregate welfare and reduce spatial inequality at the same time.

RESEARCH PAPERS

"(Trade) War and Peace: How Can International Sanctions Be Imposed Most Cost Efficiently?", with Gustavo de Souza, Haishi Li, and Yuan Mei, 2023. R&R Journal of Monetary Economics.

Trade sanctions are a common instrument of diplomatic retaliation. To guide current and future policy, we ask: What is the most cost-efficient way to impose trade sanctions against Russia? To answer this question, we build a quantitative model of international trade with input-output connections. Sanctioning countries simultaneously choose import tariffs to maximize their welfare (measured with real income) and to minimize Russia’s welfare, with different weights placed on these objectives. We find, first, the sanctioning countries can cause moderate economic damage in Russia, with Russian welfare falling 1.3% to 2.9%, depending on whether Russia retaliates or not. Second, for countries with a small willingness to pay for sanctions against Russia, the most cost-efficient sanction is a uniform, about 20% tariff against all Russian products. Third, if the European Union (EU) is willing to pay at least US$0.67 for each US$1 drop in Russian welfare, an embargo on Russia’s mining and energy sector products and about 50% tariffs on all other imports from Russia is the most cost-efficient policy. Finally, if countries target politically relevant sectors, a global embargo against Russia’s mining and energy sector is the cost-efficient policy even when there is a small willingness to pay for sanctions.

“Tariffs as Bargaining Chips: A Quantitative Analysis of US-China Trade War”, with Yuan Mei and Tong Ni

The Biden administration maintains Trump tariffs on Chinese imports, contrary to Biden's campaign commitments. We investigate the hypothesis that these tariffs serve as a leverage in future trade talks with China. Our quantitative model, incorporating disaggregated U.S. regions and international trade linkages, estimates bargaining power and simulates tariff bargaining outcomes. Results show consistent post-trade war negotiation improvements in U.S. welfare regardless of bargaining power. With an estimated U.S. bargaining power of 0.47, the post-war negotiation yields additional 0.04% gains for U.S. taking the trade war impacts into account.

Professor CHANG Pao Li
Email: plchang@smu.edu.sg
Tel: +65 68280830

Professor MEI Yuan
Email: yuanmei@smu.edu.sg
Tel: +65 68085212

Professor MA Lin
Email: linma@smu.edu.sg
Tel: +65 68280876

WORKING PAPERS

"Using Satellite-observed Geospatial Inundation Data to Identify the Impacts of Flood on Firm-level Performances: The Case of China during 2000--2009", With Pao-Li Chang (Job Market Paper)

Among the first in the literature, this paper combines high-resolution satellite-observed inundation maps with geocoded firm-level data to identify the flood exposure at the firm level. We apply the methodology to study the impact of floods on micro-level firm performances in China for the period 2000-2009. Being hit by a flood is associated with an annual loss of output and productivity of around 6% and 5%, respectively, which persists in the long run. The effects are heterogeneous across types of firms and locations of the floods. Firms that are tangible-asset intensive are more negatively affected by the flood events. Meanwhile, the effects on firms located in flood-prone counties are less severe and shorter-lived, suggesting better adaptation of firms experienced with floods. The impacts of floods extend to non-inundated firms in surrounding areas (of 4 kilometres in radius), but the negative effects are much smaller (2% on average) and diminish after three years. Firms beyond the immediate neighborhood expand their output from the second year onward, in contrast with the permanent shrinkage of the inundated firms. By aggregating the firm-level data to the county level, we further identify negative effects of floods at the extensive margin: the firm exit (entry) rate is higher (lower) in counties that are hit by floods, and the effects are stronger in counties subject to more severe floods.

"The Response of the Chinese Economy to the U.S.-China Trade War: 2018-2019", With Pao-Li Chang and Kefang Yao

In this paper, we follow the micro-to-macro approach of Fajgelbaum et al. (2020) and analyze the impacts of the 2018-2019 U.S.-China trade war on the Chinese economy. We use highly disaggregated trade and tariff data with monthly frequency to identify the demand/supply elasticities of Chinese imports/exports, combined with a general equilibrium model for the Chinese economy (that takes into account input-output linkages, and regional heterogeneity in employment and sector specialization) to quantify the partial and general equilibrium effects of the tariff war. This complements the studies focused on the ex post response of the U.S. economy by Amiti et al. (2019), Flaaen et al. (2020), Fajgelbaum et al. (2020), and Cavallo et al. (2021).

Professor SU Liangjun
Email: sulj@sem.tsinghua.edu.cn
Tel: +86 10 62789506

Professor Yichong ZHANG
Email: yczhang@smu.edu.sg
Tel: +65 68280881

Professor Peter C. B. PHILLIPS
Email: peter.phillips@yale.edu
Tel: (203) 432-3695

WORKING PAPERS

"Efficient Nonparametric Estimation of the Generalized Panel Data Transformation Models with Fixed Effects", (Job Market Paper)

In this article, we consider a generalized panel data transformation model with fixed effects where the structural functions are assumed to be additive. Our model does not impose parametric assumptions on the transformation function, the structural function, or the distribution of the idiosyncratic error term. We propose a multiple-stage Local Maximum Likelihood Estimator (LMLE) for the structural functions. In the first stage, we apply the regularized logistic sieve method to estimate the sieve coefficients associated with the approximation of a composite function and then apply a matching method to obtain initial consistent estimators of the additive structural functions. In the second stage, we apply the local polynomial method to estimate certain composite function and its derivatives to be used later on. In the third stage we apply the local linear method to obtain the refined estimator of the additive structural functions based on the estimators obtained in Steps 1 and 2. The greatest advantage is that all minimization problems are convex and thus overcome the computational hurdle for existing approaches to the generalized panel data transformation model. The final estimates of the additive terms achieve the optimal one-dimensional convergence rate, asymptotic normality and oracle efficiency. The Monte Carlo simulations demonstrate that our new estimator performs well in finite samples.

"Efficient Nonparametric Estimation of the Generalized Additive Model with an Unknown Link Function"

In this article, we consider a generalized additive model with an unknown link function (GAMULF). Our model does not impose parametric assumptions on the link function or the distribution of the idiosyncratic error term. We propose a three-stage nonparametric least squares (NPLS) estimation procedure for the additive functions. In the first stage, we estimate conditional expectation by the local-linear kernel regression and then apply matching method to the splines series to obtain initial estimators. In the second stage, we use the local-polynomial kernel regression to estimate the link function. In the third stage, given the estimators in Stages 1 and 2, we apply the local linear kernel regression to refine the initial estimator. The great advantage of such a procedure is that the estimators obtained at all stages have closed-form expressions, which overcomes the computational hurdle for existing estimators of the GAMULF model. The final estimators of the additive terms achieve the optimal one-dimensional convergence rate, asymptotic normality and oracle efficiency. Monte Carlo simulations demonstrate that our new estimator performs well in finite samples.

Professor Yichong ZHANG
Email: yczhang@smu.edu.sg
Tel: +65 68280881

Professor SU Liangjun
Email: sulj@sem.tsinghua.edu.cn
Tel: +86 10 62789506

Email: peter.phillips@yale.edu
Tel: (203) 432-3695

Professor Jia LI
Email: jiali@smu.edu.sg
Tel: +65 68280890

WORKING PAPERS

"Panel Data Models with Time-Varying Latent Group Structures",  with Peter C.B. Phillips and Liangjun Su (Job Market Paper)

This paper considers a linear panel model with interactive fixed effects such that individual heterogeneity is captured by latent group structure and time heterogeneity is captured by an unknown structural break. We allow the model to have different numbers of groups and/or different group memberships before and after the break. With the preliminary estimates by nuclear norm regularization followed by row- and column-wise linear regressions, we estimate the break point based on the idea of binary segmentation and the latent group structures together with the number of groups before and after the break by sequential testing K-means algorithm simultaneously. We show that the break point, the number of groups and the group membership can be estimated correctly with probability approaching one. Finite sample performance of the methodology is illustrated via Monte Carlo simulations and a real dataset application.

"Low-rank Panel Quantile Regression: Estimation and Inference",  with Liangjun Su and Yichong Zhang (Submitted to Annals of Statistics)

In this paper, we propose a class of low-rank panel quantile regression models which allow for unobserved slope heterogeneity over both individuals and time. We estimate the heterogeneous intercept and slope matrices via nuclear norm regularization followed by sample splitting, row- and column-wise quantile regressions and debiasing. We show that the estimators of the factors and factor loadings associated with the intercept and slope matrices are asymptotically normally distributed. In addition, we develop two specification tests: one for the null hypothesis that the slope coefficient is a constant over time and/or individuals under the case that the true rank of slope matrix equals one, and the other for the null hypothesis that the slope coefficient exhibits an additive structure under the case that the true rank of slope matrix equals two. We also illustrate the finite sample performance of estimation and inference via Monte Carlo simulation and real datasets.

WORK IN PROGRESS

"Inference for Quantile Factor Models"

Professor Tomoki FUJII
Email: tfujii@smu.edu.sg
Tel: +65 68280279

Professor HUANG Fali
Email: flhuang@smu.edu.sg
Tel: +65 68280859

Professor Madhav Shrihari ANEY
Email: madhavsa@smu.edu.sg
Tel: +65 68280644

WORKING PAPERS

"Confucian Literati and Long-run Development in Northern Vietnam",  with Tomoki Fujii (Job Market Paper)

Following Max Weber’s thesis, studies have suggested that Confucianism could impede economic growth despite its positive effects. In this paper, we revisit the impact of Confucianism on long-run development in a less explored context – northern Vietnam. Using the variation in Confucian literati across 217 historical districts between the Primitive Le and Nguyen dynasties (1426-1919), we find that districts with a greater exposure to Confucianism have experienced better economic outcomes over the past century. The result is robust to controlling for a battery of confounders and using the distance to exogenously located hermit scholars as an instrument. We show that the impact of Confucianism can be attributed to the persistence of a culture of respect for education and norms of collective action, which facilitated human and social capital accumulation, public goods provision, and economic transition.

"Sadder but Wiser: Effects of Natural Disasters on Long-term Orientation",  with James B. Ang, Per G. Fredriksson, and Jun Wang

This study attempts to uncover the origin of long-term orientation (LTO) by exploiting the variation in the historical exposure to natural disasters. We combine data on significant natural disasters occurring during 1500-1980 with LTO data derived from multiple World Values Survey and European Values Survey waves covering years 1981-2020, yielding a sample of around 361,000 respondents. We test two opposing hypotheses borrowed from psychology. We find robust evidence at the individual-, subnational-, and country-level in favor of the view that a larger exposure to catastrophic events increases the degree of LTO. This finding holds up to using alternative samples and measures, including geo-climatic confounders, and controlling for the influence of institutions and other cultural traits. The psychological process of future-oriented coping is a mechanism that helps explain our results. To cope with adverse events, individuals undertake cognitive and behavioral actions that prioritize the long term, thus shedding light on how LTO is linked to disaster exposure.

WORK IN PROGRESS

"Clan Network and Market Integration in Qing China"

Professor Jun YU
Email: yujun@smu.edu.sg
Tel: +65 68280858

Professor Jia LI
Email: jiali@smu.edu.sg
Tel: +65 68280890

Email: peter.phillips@yale.edu
Tel: (203) 432-3695

Professor Weikai LI
Email: wkli@smu.edu.sg
Tel: +65 68289651

WORKING PAPERS

"Alternative Parametric Models for Spot Volatility in High Frequency: A Bayesian Approach",  with Professor Jun Yu and Professor Jia Li (Job Market Paper)

This paper proposes several alternative parametric models for spot volatility in high frequency, depending on whether or not jumps, seasonality, and announcement effects are included. Together with these alternative parametric models, nonlinear non-Gaussian state-space models are introduced based on the fixed-k estimator of spot volatility of Bollerslev, Li, and Liao (2021). According to Bollerslev, Li, and Liao (2021), the log fixed-k estimator of spot volatility equals the true log spot volatility plus a non-Gaussian random variable. Bayesian methods are introduced to estimate and compare these alternative models and to extract volatility from the estimated models. Simulation studies suggest that the Bayesian methods can accurately estimate the parameters, select the true model, and extract volatility. Empirical studies using high frequency market indexes and individual stock prices reveal several important results. As an application of extracting volatility, we quantify the strategic value of information.

RESEARCH PAPERS

"Sparse Structure of Stochastic Discount Factor in the Chinese Stock Market: A Bayesian Interpretable Machine-learning Approach"

This paper reviews a Bayesian interpretable machine-learning method proposed by Kozak, Nagel, and Santosh (2020). We show how the method can link two strands of literature, namely the literature on empirical asset pricing and the literature on statistical learning. Based on a recently developed data-cleaning technique, we obtain 123 financial and accounting cross-sectional equity characteristics in the Chinese stock market. When applying the method of Kozak, Nagel, and Santosh (2020) to the Chinese stock market, we find that it is futile to summarize the stochastic discount factor (SDF) in the Chinese stock market as the exposure of several dominant cross-sectional equity characteristics in-sample. A cross-validated out-of-sample analysis further supports this finding.

“Do Volatility-Managed Portfolios Work? Empirical Evidence from the Chinese Stock Market”

Using data from the Chinese stock market, we have found that the main empirical findings in Moreira and Muir (2017) break down. Based on the new empirical findings, we exploit a comprehensive set of 99 equity strategies in the Chinese stock market to analyze the economic value of managed portfolios. Based on these 99 equity trading strategies, we find that there exists no systematic gain from scaling the original portfolios using volatility. Our empirical results suggest that one should be careful to use volatility-managed portfolios in practice as the expected performance gains are rather limited.

WORK IN PROGRESS

“How is Fund Investment Exposed to Stock-level Characteristics? Evidence from U.S. Equity Market”, with Assistant Professor Xiaobin Liu and Associate Professor Tao Zeng

This paper documents how the Instrumented Principal Component Analysis (IPCA) is applied in uncovering the driving factors associated with firm-level characteristics to which fund managers holding these stocks are exposed. IPCA is a specific statistical learning methodology featuring both latent factor structure and dynamic factor loading which accordingly can simultaneously handle dimensionality and time-varying parameter concerns for financial econometric modeling. Linear structure is retained and therefore the corresponding statistical hypothesis testing is possible to be implemented. In this paper, we first construct fund-level indices as the measure of exposure of each fund to firm-level characteristics (commonly referred to as "anomalies'' in accounting or finance literature) and document the empirically stylized facts revealed from the constructed dataset. With the constructed fund-level index, we apply the novel IPCA methodology along with our proposed regularized IPCA to discuss how the mutual fund returns are exposed to characteristics of managed assets (specifically those firm-level characteristics of assets held by a fund).

“Estimating Expected Return Function Nonparametrically: Based on BART”

This paper documents the empirical Implementation of estimating the expected return function nonparametrically using the Bayesian Additive Regression Tree (BART) method. Within this newly introduced nonparametric framework, general non-linearity is allowed for the specification of the model when the dimension of covariates used for prediction is large and the underlying non-linear relationship is hard to detect. By applying BART, we document which firm-level characteristics should be adopted as the most influential predictors for estimating expected return and the out-of-sample performance of BART for prediction as well. I have also extended the whole framework to the China stock market and the global financial market for empirical comparison. Our finding suggests that (i) the performance of BART approximates the results obtained from neural-network, which is a specific machine-learning method documented with dominating out-of-sample prediction performance; (ii) The machine-learning-based method (specifically BART) surely outperforms the benchmark linear model, but in terms of investment strategy constructed from prediction, there is not much significant difference between machine-learning methods and linear benchmark; (iii) China stock market is relatively more predictable in comparison to the U.S. stock market in terms of out-of-sample prediction-accuracy measure.