Tree-Based Models for the Efficient Valuation of Large Variable Annuity Portfolios

32 Pages Posted: 28 Sep 2018

See all articles by Zhiyu Quan

Zhiyu Quan

University of Connecticut - Department of Mathematics

Guojun Gan

University of Connecticut - Department of Mathematics

Emiliano A. Valdez

University of Connecticut - Department of Mathematics

Date Written: September 10, 2018

Abstract

Because of attractive guarantee features, variable annuities have become popular retirement and investment vehicles. Insurance companies often face the challenge of managing the financial risks associated with these guarantees. Dynamic hedging is a common practice among insurers but requires frequent valuation of the fair market values of the guarantees. Insurers rely on the use of Monte Carlo simulation, which is flexible but computationally intensive. Metamodels are increasing in popularity as efficient approaches for addressing the computational issues. In this paper, we use a synthetic dataset with 190,000 variable annuity contracts to empirically examine the performance of tree-based models for the valuation of the guarantees. In particular, we compare the predictive accuracy of five tree-based models that include traditional regression tree, ensemble methods, and trees based on unbiased recursive partitioning, as well as models based on ordinary kriging and GB2 regression. Our results indicate that tree-based models are generally very efficient in producing more accurate predictions and the gradient boosting ensemble method is considered the most superior.

Keywords: variable annuity, portfolio valuation, metamodeling, tree-based models

JEL Classification: A10, C10, C15, C55, C90, G22

Suggested Citation

Quan, Zhiyu and Gan, Guojun and Valdez, Emiliano A., Tree-Based Models for the Efficient Valuation of Large Variable Annuity Portfolios (September 10, 2018). Available at SSRN: https://ssrn.com/abstract=3247100 or http://dx.doi.org/10.2139/ssrn.3247100

Zhiyu Quan

University of Connecticut - Department of Mathematics ( email )

196 Auditorium Rd U-3009
Storrs, CT 06269-3009
United States

Guojun Gan

University of Connecticut - Department of Mathematics ( email )

196 Auditorium Rd U-3009
Storrs, CT 06269-3009
United States

Emiliano A. Valdez (Contact Author)

University of Connecticut - Department of Mathematics ( email )

341 Mansfield Road U-1009
Storrs, CT 06269-1009
United States

HOME PAGE: http://www.math.uconn.edu/~valdez

Register to save articles to
your library

Register

Paper statistics

Downloads
28
Abstract Views
209
PlumX