We consider forecasting a single time series when there is a large number of predictors and a possible nonlinear effect. The dimensionality was first reduced via a high-dimensional factor model implemented by the principal component analysis. Using the extracted factors, we develop a link-free...

Testing multi-factor asset pricing models is instrumental for the asset pricing theory and practice. Due to the accumulation of errors in estimating high-dimensional parameters, traditional quadratic-form tests such as the Wald test perform poorly against the sparse alternative hypothesis in the...

We propose a novel technique to boost the power of testing a high-dimensional vector $H:\theta=0$ against sparse alternatives where the null hypothesis is violated only by a couple of components. Existing tests based on quadratic forms such as the Wald statistic often suffer from low powers due...

We propose and validate a new measure of earnings quality based on a hidden Markov model. This measure, termed earnings fidelity, captures how faithful earnings signals are in revealing the true economic state of the firm. We estimate the measure using a Markov chain Monte Carlo procedure in a...

This study addresses the relation between the exit of venture capital and opportunistic behavior in financial disclosure. Specifically, I examine whether the exit of venture capital is associated with income-increasing earnings management in the IPO year and financial statement restatements...

I investigate whether earnings management in venture-capital-backed IPO firms has any consequence on reputation of venture capitalists. I document that discretionary accruals in the IPO year are negatively related to the number of new IPOs backed by the same lead venture capitalist during the...

A significant portion of CEOs in publicly-listed Chinese state-owned enterprises receive zero pay from the companies for which they work. Instead, they are paid directly by their controlling shareholder who can be the Chinese government or parent firms controlled by the Chinese government. While...

Our study documents a “Lemons” market failure of Chinese firms listed in the US in 2011 and a subsequent rebound by 2013. Our tests reveal that there was little difference in ex ante observable characteristics of fraudulent and non-fraudulent Chinese firms listed in the US prior to 2011...

We exploit exogenous variation in turnovers of China's municipal political leaders to study how political incentives affect firms' strategic tax avoidance behavior. We document robust cycles in corporate tax compliance corresponding with the timing of political turnovers. Specifically, we find...

Using a machine learning approach to process 11 million tweets posted by S&P 1500 firms from 2011 through 2016, we find that poor corporate social responsibility (CSR) performance firms tweet more about CSR activities and use tweets that are shorter, and with more passive voice and extreme tone....