Font Size: a A A

Research On Quantitative Stock Selection And Quantitative Timing Strategy

Posted on:2022-01-07Degree:MasterType:Thesis
Country:ChinaCandidate:N YuFull Text:PDF
GTID:2480306311475894Subject:Applied Statistics
Abstract/Summary:PDF Full Text Request
Quantitative trading has matured overseas,but there is still much room for development in China,which is a hot topic in China in recent years.Quantitative trading needs to consider two key dimensions:quantitative stock selection and quantitative timing and combining them can reduce risks and obtain stable returns.Therefore,in this paper,we will focus on these two dimensions.The research in this paper is mainly based on BigQuant,JoinQuant,Uqer and Nuggets quantitative trading platforms.The research range is from May 2015 to January 2020,including bull market,bear market and shock market.In the part of quantitative stock selection,firstly,the industry rotation stock selection is carried out,and six industries are selected and the best industry index is obtained to select stocks.Then,the back test is carried out and the CSI 300 Index is compared as a benchmark.However,the essence of industry rotation is the limitation of resonance or deviation between valuation and fundamentals,which makes the stock selection strategy perform better only in volatile markets;Because the stock price fluctuated too much in this time period,stock selection with characteristic volatility was carried out,mainly based on the principle that stocks with low characteristic volatility have higher expected returns in the future.We decompose the risk and use CAPM,Fama-French three factors,Carhart four factors and Fama-French five factors to measure idiosyncratic volatility.Through IC test and portfolio construction,we find that the stock selection ability of low idiosyncratic volatility factor based on Fama-French three factors is the best.After backtesting,we all outperformed the benchmark return in this period,and can get 10%excess return.In the timing part,volatility timing works well in foreign markets,so we introduce it into China market.Its principle is that there is a negative correlation between risk-return trade-off and volatility,so when volatility is low,the proportion of positions should be increased,and vice versa.We tested the effectiveness of this strategy with the data of MKT,HML and SMB factors in the domestic market,in which the HML factor failed to pass the test.Then we drew a strategic return curve and compared it with the CSI 300 index,and found that the strategic return was basically above the benchmark return,but most of the time the strategic return was near zero and overlapped with the benchmark return curve,so it would not get too much excess return.Therefore,the market performance of volatility timing strategy in China during this period was not very good,mainly because the risk-return balance and volatility in China's market were combined.In the hidden Markov model,the profit effect of the model is better,and finally it can reach about 3.6 times of the benchmark profit.Finally,the combination strategy of stock selection with low idiosyncratic volatility and HMM timing is constructed.Compared with pure stock selection strategy,it can largely avoid the impact of volatile market,improve returns and reduce risks.Then,by discussing the advantages and limitations of various strategies,combined with the conclusions,it can provide reference for investors to make decisions.
Keywords/Search Tags:Industry rotation, Special volatility stock selection, Volatility timing, Hidden Markov model
PDF Full Text Request
Related items