Bivariate Distribution Regression with Application to Insurance Data. Joint with Tatsushi Oka and Dan Zhu. Insurance Mathematics and Economics 113 (2023): 215-232.
Abstract: This article presents an estimation method for modeling the conditional joint distribution of bivariate outcomes based on the distribution regression and factorization methods, avoiding imposing global parametric assumptions across the entire distribution. In contrast to existing parametric approaches, our method can accommodate discrete, continuous, or mixed variables, and provides a simple yet effective way to capture distributional dependence structures between bivariate outcomes and covariates. In an application to the study of a motor third-party liability insurance portfolio, the proposed method effectively estimates risk measures such as the conditional Value-at-Risk and Expected Shortfall.
Valuation of variable annuities with guaranteed minimum maturity benefits and periodic fees. Joint with Meiqiao Ai, Zhiming Zhang, and Dan Zhu. Scandinavian Actuarial Journal (2023): 1-27.
Abstract: This paper focuses on the valuation of variable annuities with a guaranteed minimum maturity benefit under a regime-switching Lévy model. The model allows policyholders to surrender their annuities and receive a surrender benefit at predetermined tenor times before maturity. Additionally, we consider a state-dependent periodic fee structure where fees are deducted from the policyholder's account if it exceed a certain level at discrete time points. Incorporating this fee structure, the Fourier cosine series expansion method based on characteristic functions is employed to determine the values and optimal surrender strategies for variable annuity contracts. Finally, we provide a comprehensive set of numerical examples to demonstrate and thoroughly assess the effectiveness of our approach.
Estimating the Gerber-Shiu function under a risk model with stochastic income by Laguerre series expansion. Joint with Wen, Su, and Benxuan Shi. Communications in Statistics-Theory and Methods 49.23 (2020): 5686-5708.
Abstract: In this paper, we concern the statistical estimation of the Gerber-Shiu function under a risk model with stochastic premiums. We express the Gerber-Shiu function by Laguerre series expansion and estimate it based on observed information. Moreover, we study the convergence rate of our proposed estimator and illustrate the excellent performance of the estimator by simulations with finite sample size.
Distributional Vector Autoregression: Eliciting Macro and Financial Dependence. Joint with Tatsushi Oka, and Dan Zhu. (Submitted)
Abstract: In this study, we extend the vector autoregression framework by introducing a flexible multivariate distributional regression approach that models multivariate time series without imposing restrictive parametric distribution assumptions. We develop a distributional impulse response function that captures the future e ect of distributional disturbances in any variable within the system, providing a more detailed view of dynamic heterogeneity. We propose a straightforward estimation method and establish its asymptotic properties under weak dependence assumptions. Applying this approach to U.S. economic data, we analyze the interplay among real GDP growth, financial conditions, and interest rates. Our model exhibits strong forecasting capabilities compared to existing alternatives. By examining distributional interactions and monetary policy impacts, particularly during the Great Recession, we uncover complex macroeconomic dynamics.
Inflation Target at Risk: A Time-varying Parameter Distributional Regression. Joint with Tatsushi Oka, and Dan Zhu.
Abstract: Macro variables frequently display time-varying distributions, driven by the dynamic and evolving characteristics of economic, social, and environmental factors that consistently reshape the fundamental patterns and relationships governing these variables. To better understand the distributional dynamics beyond the central tendency, this paper introduces a novel semi-parametric approach for constructing time-varying conditional distributions, relying on the recent advances in distributional regression. We present an efficient precision-based Markov Chain Monte Carlo algorithm that simultaneously estimates all model parameters while explicitly enforcing the monotonicity condition on the conditional distribution function. Our model is applied to construct the forecasting distribution of inflation for the U.S., conditional on a set of macroeconomic and financial indicators. The risks of future inflation deviating excessively high or low from the desired range are carefully evaluated. Moreover, we provide a thorough discussion about the interplay between inflation and unemployment rates during the Global Financial Crisis, COVID, and the third quarter of 2023.
MIDAS Regression Tree: Understanding Macro/Financial Effect on Mortality Movement
Climate Insurance with Panel Distributional Regression
Building Conditional Distributions via Monotonic Distributional Regressions