Research On The Strategy Of Pairing Conversion Arbitrage Of Classification Fund

Classification fund belongs to a kind of structured products,starting from abroad.Under the new economic system in China,China's financial institutions draw on the establishment and operation of foreign classification funds,and combine with China's financial system and the basic national conditions,the development of a new innovative products.Since July 2007 Chinese first grade fund,UBS SDIC Ruifu classification of stock issuance has been established for ten years,many investors have become popular products.By the end of the two quarter of this year,the total size of the B fund reached 61 billion 300 million copies,and the total share of the circulation reached 47 billion 856 million copies.Although the new regulations,public funds were prohibited from issuing new product classification,the threshold of investors is also rising,but in the long term,which is conducive to purify the environment of the classification of the fund market,the elimination effect is poor,the performance index tracking theme repeated poor mobility products.After stabilizing the market environment,the classification fund market is expected to be further standardized.These measures also show the determination of China's regulators to do a good job of grading fund market.At present,there are many arbitrage strategies for classified funds,and the market focuses on the design of transaction arbitrage trading strategies.At present,there are hedging strategies for futureshedging and arbitrage strategies of holding bottom positions in some common trading strategies.But the arbitrage strategy itself arbitrage influence factors to explore deficiencies.This paper is divided into six chapters.The first chapter is the introduction,which mainly introduces the related research background,research status,methods and purposes,and the framework.In the second chapter,we introduce the development and classification of the classification fund.The third chapter and the fourth chapter uses the method of empirical analysis,firstly establish the factor system,according to the screening factor by using multiple linear regression method and nonlinear depth of machine learning to explore which factors of paired conversion premium discount arbitrage mechanism will affect the result of arbitrage.Finally,it is concluded that the two classifier of nonlinear neural network is very effective for the forecast of arbitrage.