An Empirical Analysis Of Quantitative Strategies In Securities Market And Construction Of Transaction Model

Quantitative strategy analysis system in the Stock Market is from the use of computers and the popularity of big data.It has the following characteristics: First,it is objective transaction, avoiding interference from people’s subjective emotions, to achieve a smooth and stable growth rate in funds. Second, quantitative strategy analysis can provide detailed statistics.Through a long period backtesting in date for each king of investment strategy, you can identify the effectiveness and will avoid the long-term investment which will cause losses. Third, quantitative strategy analysis can search for the optimization strategy for a particular kind of investment stragety through data validation, building a new trading method more usefulthan the original investment strategy.The aim of this paper is to use quantitative methodsbased on traditional analysis of Morning Starby using empirical research, including construction, evaluation, optimization, and eventually build a set of quantitative stragety to evaluate the results which is better than the original transation model and traditional transaction model.This paper is divided into six parts: the first part is the introduction, introduces the background and significance of the research, professional practice and literature review; the second part is summary, including introduction, basic structure and testing method of quantitative stragety; the third part is the scheme of empirical research, by using TongDaxinLanguage as the foundation, through the provisions of the evaluation target sample, study time period, entrance and exit point of transaction and optimization plan and thus to construct optimization goalsto provide theoretical guidance for the optimization; the fourth part is the empirical research records, each set contains model detection source code and parameter optimization scheme whichbelongs to the empirical research, and finally detect the correlation of all detection conditions with the winning rate; the fifth part is the part of building a better traction model, by using othe fourth Part of the data, to build up a set of high winning trading model.The final result is to construct a trading stragety which winning percentage and total profit beyond the traditional model. The maximum loss of the model is less than the traditional investment model, and the relative yield of the model during the investment period is better than the same period of the market index.