主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

Chinese Journal of Management Science ›› 2023, Vol. 31 ›› Issue (7): 60-67.doi: 10.16381/j.cnki.issn1003-207x.2021.0239

• Articles • Previous Articles     Next Articles

Nonparametric Option Pricing under Multivariate No-arbitrage Constraints

LI Qing1, GE Xiang-yu1, XIANG Xiu-li2   

  1. 1. School of Statistics and Mathematics, Zhongnan University of Economics and Law, Wuhan 430073, China;2. School of Economics, Wuhan Textile University, Wuhan 430200, China
  • Received:2021-02-02 Revised:2021-05-08 Online:2023-07-17 Published:2023-07-17
  • Contact: 向秀莉 E-mail:xlxiang@wtu.edu.cn

Abstract: This existing nonparametric option pricing models under no-arbitrage constraints only considered the constraint of strike price along a single maturity, this paper will incorporate all the no-arbitrage constraints, such as maturity, underlying asset price, risk-free rate, and volatility. It’s difficult to estimate the nonparametric model under multivariate no-arbitrage constraints, we will reduce the dimension by variable transformation, and the multivariate will be reduced to one dimension. The number of option strikes contracts is infinite in single maturity, the problem can be solved by indexed multiple time-to-maturities options portfolios with variable transformation. Then, the nonparametric option pricing model under no-arbitrage constraints can be estimated efficiently using a quadratic programming algorithm, which will be computed easily in soft. Finally, we make empirical analysis by SSE 50 ETF options data, and make conclusion that the pricing effect of nonparametric option pricing model under multivariate no-arbitrage constraints is better than Black-Scholes model.

Key words: No-arbitrage Constraints; Multi-term Portfolio; Multivariate Shape Constraints; Local Polynomials Regression; Quadprog Programing

CLC Number: