The Quantitative Investment Model And Its Strategy’s Realizing Based On Survival Analysis

With the popularity of quantitative investment in China, it’s inexorable for using mathematical methods and techniques to quantify the phenomena in financial market. Conforming to the trend of the time, this article creatively proposes survival model to describe the survival characters of stocks’ successive rises or declines. Additionally, it improves conventional test methods to the efficiency of technical indices and quantify their level of efficiency. Lastly, depending on them, it puts forward a new investment strategy and evaluates the performance of the strategy.When it comes to testing the efficiency of technical indices, this article does the single technical index efficiency analysis through constructing MACD, RSI and OBV index, regarding the close price of CITICS as the example, and finds that MACD and RSI has strong ability in following the trend of this stock so that they are somewhat valuable to the investors, while OBV has little, even none. What’s more, through analysis of combining the effective trade signal from different technical indices, it’s found that the function of strengthen each other does exist among effective indices. So it infers that we need analyze more indices in order to find more effective indices so as to provide the investors more scientific system in implement of technical indices.When it comes to quantifying technical indices’ level of efficiency, this article does it through Cox model. According to the results, it creatively quantifies the level of efficiency of MACD and RSI, realizing the comparison among different indices. It shows that MACD’s level of efficiency is2.29 while the RSI’s is 2.11. Moreover, RSI is obviously better than MACD in the yield rate of successive decline.Through the model above describing the phenomenon of one stock’s successive rises or declines, this article constructs a quantitative investment strategy which can be classified into the time-selecting strategies. First, this article structures the time-dependent covariate Cox model and take its parameters into the statistical test; Secondly, the investment strategy is structured according to the result from model, mainly about the condition of openning positions and closing positions; Thirdly, the indecies measuring every aspect of the strategy are observed to conclude whether it’s workable; Last, more underlying assets are taken into the backtest using the same strategy, so that observe whether it’s stable in other markets. From the resulting information ratio, although this strategy can’t always exceed the buy and hold strategy, it’s worthy enough to probe deeply from the indecies in so many underlying assets, mar ratio and cox-beta which is put forward creatively in this article.This article aims to put forward a new method of quantitative investment to study the security market, seeing the successive rises or declines of underlying assets as the basic units, from verifying the effectiveness of the trading signals of time-choosing strategies or technical indecies to quantify their levels of efficiency, even to structure the model or strategy combining different signal. Apart from signals of strategies, this method can be generalized to many other field, like evaluating the persistence of the funds’ performance.