Research

PUBLICATIONS

1."Bootstrap inference for pre-averaged realized volatility based on non-overlapping returns," with Sílvia Gonçalves and Nour Meddahi, 2014, Journal of Financial Econometrics, 12(4), 679-707.

2."Validity of Edgeworth expansions for realized volatility estimators," with Bezirgen Veliyev, 2016, Econometrics Journal, 19(1), 1-32. This paper receives the Denis Sargan Econometrics Prize for 2016, the Royal Economic Society.

3."Bootstrapping pre-averaged realized volatility under market microstructure noise," with Sílvia Gonçalves and Nour Meddahi, 2017, Econometric Theory, 33 (4), 791-838.

4."Bootstrapping integrated covariance matrix estimators in noisy jump-diffusion models with non-synchronous trading," 2017,  Journal of Econometrics, 197(1), 130-152.

5."A local stable bootstrap for power variations of pure-jump semimartingales and activity index estimation," with Rasmus T. Varneskov,  2017,  Journal of Econometrics, 198(1), 10-28.

6."Is the diurnal pattern sufficient to explain the intraday variation in volatility? A nonparametric assessment," with Kim Christensen, and Mark Podolskij, 2018,  Journal of Econometrics, 205(2), 336-362.

7."Bootstrapping high-frequency jump tests," with Prosper Dovonon, Sílvia Gonçalves, and Nour Meddahi, 2019,  Journal of American Statistical Association, 114,  793-803.  Online Appendix.

8."A local Gaussian bootstrap method for realized volatility and realized beta," 2019,  Econometric Theory, 35, 360-416.

9."The local fractional bootstrap," with Mikkel Bennedsen, Asger Lunde, and Mikko Pakkanen,  2019, Scandinavian Journal of Statistics, 46, 329-359. PDF.

10."Inference for local distributions at high sampling frequencies: a bootstrap approach," with Rasmus T. Varneskov,  2020, Journal of Econometrics,  215(1), 1-34. (Lead Article).

11."Estimating the variance of a combined forecast: bootstrap-based approach," with Kajal Lahiri, 2023, Journal of Econometrics, 232(2), 445-468. 

12."A wild bootstrap for dependent data," 2023, Econometric Theory, 39(2), 264-289 .

13." Are Some Forecasters Really Better Than Others? A Further Note," with Kajal Lahiri, 2023, Journal of Money, Credit and Banking, 55, 577-593.

14."Bootstrapping two-stage quasi-maximum likelihood estimators of time series model", with Silvia Goncalves, Andrew Patton and Kevin Sheppard, PDF.  MATLAB Toolbox, 2023, Journal of Business and Economics Statistics, 41(3), 683-694. 

15."Bootstrapping Laplace Transforms of Volatility," with Rasmus T. Varneskov, and  Zhi Liu., 2023,  Quantitative Economics, 14(3),  1059-1103.

WORKING PAPERS

 1."A Nonparametric test for commonality in intraday high-frequency data," with Kim Christensen and Zhi Liu, 2023.

Abstract:  The association between log-price increments of financial assets, as measured by their spot correlation coefficient estimated from high-frequency data, exhibits a pronounced upward-sloping and almost piecewise linear relationship at the intraday horizon. There is notably lower (on average less positive) correlation in the morning than in the afternoon. We develop a nonparametric infill asymptotic theory to detect the presence of such systematic features in a correlation process. The proposed test statistic has a limiting standard normal distribution under the null hypothesis of no diurnal pattern in correlation, whereas it diverges and thus has power under an alternative with deterministic variation in the correlation. We run a Monte Carlo simulation to discover the finite sample properties of the test statistic, which are close to the large sample predictions, even for very small sample sizes and realistic levels of diurnal correlation. Moreover, the test statistic is found to be robust against stochasticity in the intraday correlation process. In practice, the test statistic rejects about one-third of days, when high-frequency data from single stocks are paired with a proxy for the market portfolio. In a hedging experiment, we show there are nontrivial gains associated by accommodating diurnal variation in the evolution of the correlation coefficient.

2."Reliable Wild Bootstrap Inference with Multiway Clustering," with Jiahao Lin, 2023.

Abstract: This paper studies wild bootstrap-based inference for regression models with multiway clustering. Our proposed method is a multiway counterpart to the (one-way) wild cluster bootstrap approach introduced by Cameron et al. (2008). We establish the validity of our method for studentized statistics. Theoretical results are provided, accommodating arbitrary serial dependence in the common time effects, an aspect excluded by existing two-way bootstrap-based methods. Simulation experiments document the potential for enhanced inference with our novel approach. We illustrate the effectiveness of the method by revisiting empirical studies involving multiway clustered and correlated data.

3."Can mutual fund "stars" really pick stocks? New evidence from a wild bootstrap analysis," with Jiahao Lin, 2023.

Abstract: This paper introduces a novel approach called wild bootstrapping for analyzing mutual fund performance. Our proposed method preserves various characteristics of mutual fund databases, including entry/exit points for each fund (i.e., missing data) and cross-sectional information. We show that our proposed bootstrap tests have a near-optimal size and exhibit greater power compared to widely used standard bootstrap methods for evaluating mutual fund performance. Additionally, we present a novel approach to picking mutual funds that do not underperform others. Our empirical results indicate that a measurable fraction of funds outperform the market. Furthermore, we extend our methods to assess mutual fund market timing abilities.

4."Heterogeneity in Carbon Intensity Patterns : A Subsampling Approach," with Johnson Kakeu and Li Lu, 2023.

Abstract: Carbon intensity, defined as carbon dioxide (CO₂) emissions per unit of gross domestic product (GDP), is a critical metric for assessing the effectiveness of climate policy across nations. This paper presents an analysis of the persistence and stationarity of carbon intensity data across 176 countries and 44 regions from 1990 to 2014, employing subsampling confidence intervals. Subsampling is a robust statistical technique that performs well with finite samples and requires minimal assumptions about the data. Our findings categorize countries into three distinct groups based on their carbon intensity patterns: convergent, persistent, and divergent. We observe that climate mitigation policies in countries with a convergent pattern tend to have only temporary effectiveness, whereas in countries with a divergent pattern, such policies can lead to permanent changes. Additionally, using unsupervised learning methods, we delve into the underlying factors influencing these classifications. This study is particularly significant for understanding the long-term impacts of climate policies, offering valuable insights for policymakers and international bodies. By identifying and analyzing these distinct patterns, our research contributes to the strategic planning and implementation of more effective and sustainable climate policies globally, aligning with the goals of international agreements like the Paris Accord.

5."Forecasting economic time series in presence of weak factors: multiple supervised learning-based approach," with Zhendong Li, 2023.

Abstract: This paper proposes an innovative supervised learning technique for dimension reduction and forecasting financial and macroeconomic time series in the presence of weak factors that can undermine the effectiveness of traditional Principal Component Analysis (PCA). This approach employs double or multiple supervised learning procedures and is termed `Supervised Scaled Principal Component Analysis' (SsPCA). Initially, each predictor is scaled using its predictive slope on the target forecast variable, giving more weight to those with stronger predictive abilities and less to those with weaker ones. Subsequently, utilizing the scaled predictors, we intelligently identify the most informative subset for prediction. This involves iterative steps of supervised selection, factor extraction through PCA, and projection. The integration of a pre-selection step, aimed at choosing `targeted predictors' prior to executing the SsPCA procedure, especially using soft thresholding of supervised learning methods like elastic net, significantly enhances the predictive capacity of SsPCA. Additionally, incorporating the possibility of non-linear relationships between predictors and factors often leads to supplementary improvements. Through extensive Monte Carlo simulation exercises, we demonstrate that our proposed SsPCA procedure consistently outperforms standard PCA. Real-world examples involving macroeconomic and financial asset pricing forecasting further affirm that SsPCA generally exhibits superior performance.

6."A modified wild bootstrap procedure for Laplace transform of volatility," with Zhi Liu and Rasmus T. Varneskov, 2023.

Abstract:  In this note, we propose a modified wild (MW) bootstrap-based procedure for the realized Laplace transform (RLT) of volatility. We establish its first-order asymptotic validity

7."Misspecification-robust bootstrap t-test for irrelevant factor in linear stochastic discount factor models,"with Antoine Djogbenou, 2024.

Abstract: This paper examines the applicability of the bootstrap approach to test for irrelevant risk factors that are potentially useless in misspecified linear stochastic discount factor (SDF) models. In the literature, the misspecification-robust inference with useless factors is known to give rise to nonstandard limiting distributions bounded stochastically to compute critical values. We show how and to what extent the wild bootstrap yields a more accurate approximation of the distribution of t-statistics when testing for an unpriced factor in the context of linear SDF models. Simulation experiments and empirical tests are also used to document the relevance of the bootstrap method. 

8."Heteroskedasticity and Autocorrelation Robust Bootstrap Test for Cross-Sectional Correlation in Large Panel Models," with Chihwa Kao and Min Seong Kim, 2023.

Abstract:  We propose a novel cross-sectional correlation test for panel regression disturbances designed to maintain robustness in the presence of serial dependence. While serial dependence is a prevalent characteristic of panel data, most of the existing tests in this context are established under the assumption of serial independence. Nevertheless, these tests can lead to significant size distortion when the assumption of serial independence does not hold, as emphasized by Bertrand, Duflo, and Mullainathan (2004). To address this issue, we propose a robust test for serial dependence based on the cross-sectional dependence (CD) test introduced by Pesaran (2004, 2015). Initially, we propose the test statistic by applying studentization to the CD statistic, utilizing the series heteroskedasticity and autocorrelation robust (HAR) variance estimator. We show that our statistic is asymptotically t distributed under the so-called fixed-K asymptotics. To enhance the finite sample properties of our test, we develop a wild bootstrap procedure that replicates the null distribution, facilitating the derivation of bootstrap-based critical values. We establish the validity of our bootstrap by showing that the bootstrap statistic is consistent for the null distribution irrespective of the presence of serial and/or cross-sectional dependence under the fixed-K asymptotics. Through simulations, we demonstrate the strong performance of our test in finite sample settings.