Study On Quantitative Stock Selection Strategy Based On Growth At Reasonable Price Model

This article uses the knowledge and theory of finance,statistics,computer science and other related aspects to deeply analyze the internal mechanism of the GARP model,and finally carries out empirical analysis.The empirical model of this paper has been innovative on the basis of the past,different from the traditional quantitative stock selection model relying on human judgment,using the decision tree algorithm in machine learning to filter factors that meet the definition of GARP.The final composition of the GARP model has finished.The empirical results show that regardless of the backtest during the sample period from January 2012 to December 2014,or the sample backlog from January 2015 to October 2017,the stock portfolio selected by the GARP model can won the CSI 300 Index,which beats the benchmark reference.Among them,during the sample period,the strategy achieved a total return of 381.4%,and its annualized rate of return was 69%,which was much higher than the 53.7% total return rate of the CSI 300 index on the reference basis,and its corresponding annualized return The rate is only 15.4%.Judging from the indicators of comprehensive evaluation of earnings risks,the Sharpe index of the strategy is 2.506,which is much higher than the index of 0.672.This shows that our strategy pays a unit of risk,and the return on earnings is 2.5 times,while the index of the same period Only 0.67 times.For the backtesting outside the sample,we have increased the risk control conditions and stock options,in order to be closer to the real trading environment.With the addition of risk control conditions and stock options,the overall yield of the strategy increased by 6.1% to 120.4%,while the corresponding annualized rate of return rose by 1.3% to 32.3%,indicating that the timing conditions The stop loss condition does indeed optimize the return level of the original combination.It is worth noting that after adding the stop loss condition and stock option timing optimization condition,the maximum retracement rate of the strategy has successfully dropped from 39.8% to 12%,which not only greatly reduces the risk of loss in the portfolio,but also complies with our risk loss is lower than The 20% threshold is expected.Judging from the comprehensive evaluation indicators of benefits and risks,Sharpe index,which we value more,has risen from 0.905 below 1 to 1.534,indicating that the optimized strategy risks and benefits have been significantly improved.At the end of the article,taking the empirical results as the starting point,combining China's capital market operation to give specific research conclusions,and put forward the outlook.