Quantitative finance has been well developed in western countries. The majority trading (over 70%) have been implemented in quantitative approach. However, the choice of good algorithms/models has caused practitioners unexpected challenges during the crisis time. Quantitative trading is just in its initial developing stage in China due to the immature finance market and limited products to trade (The 1st quantitative fund just started to trade in China from 2009). Hence the study of model selections in quantitative finance for Chinese markets has both great scientific and economy values. The expected outcome will provide criteria of models selections for trading and investment in Chinese finance market.
Topics on model selection in quantitative finance involve analysing and studying partial observed big data from finance market. Various analysis methods such as data mining, computational statistics and machine learning will be used to study the observed big data. It is innovative multidisciplinary research areas, including finance, machine learning techniques and applied mathematics, which fit the goal of IDIC/CDT very well.
Model selection mainly focuses on distinguishing various competing models and picking the most appropriate model which could fit the observed data. There are various analytical methods that have been designed in model selections. Hence, I will mainly study the applications of the different criteria and computational methods regarding the models in quantitative finance. The objective of my research is to set up an effective scheme to identify a feasible criterion for specific financial data set
Despite the enormous advances in inferential methodology, there has been relatively little activity in the area of model assessment. We will apply computational approach and MCMC tools to compare different deterministic and stochastic models to a high extent to fit the observed finance data.
This work was carried out at the International Doctoral Innovation Centre (IDIC). The authors acknowledge the financial support from Ningbo Education Bureau, Ningbo Science and Technology Bureau, China's MOST, and the University of Nottingham. The work is also partially supported by EPSRC grant no EP/L015463/1.