导　　师： 梁满发

学科专业： 070103

授予学位： 硕士

作　　者： ;

机构地区： 华南理工大学

摘　　要： 证券技术分析是广泛采用的一种证券投资分析方法。投资者根据技术指标预测股市未来的行情走势以期获得更多的收益，此时，最关键的就是寻找股价反转点。如果能够较早的判断股价的反转点，做到低吸高抛，投资者将获得丰厚的回报。交易量和股价（简称量价）间的关系在技术分析中占据着重要的地位。基于量价，本文设计了一种识别反转点的模型，对于投资者进行更好的投资决策具有参考意义。 根据证券技术分析的分析理念，及收盘价和交易量间存在的协整关系，本文将收盘价和交易量同时进行考虑，设计了双指标极小投资模型（简称双极模型）。双极模型中有两个参数，分别是P和Q，其中，P代表当日收盘价是连续P天的极小值，Q代表当日交易量是连续Q天的极小值，不同的参数组合代表不同的投资模型。双极模型的设计思路是给定P和Q，若当日收盘价是连续P天的极小值并且当日交易量是连续Q天的极小值，则双极指标值取为1，否则取为0，当双极指标值由1变为0时，则买进股票；买进股票后，当股价从一个新高位置下跌，从这个新高位置开始，若4天内回落达到5/%，则卖出股票。对于股票收盘价序列和交易量序列一阶单整，且差分后序列一阶自相关的情况，理论上证明了双极模型设计的合理性。 假定投资期限为n年，利用双极模型进行投资模拟。首先，计算不同参数组合下每次买卖的投资收益，按复利计算其n年投资总收益，并计算其年均收益率。其次，根据EGARCH模型，算出每次投资的日VaR值，从而得到一组VaR序列，将其99/%分位数作为投资期限内的总风险值。用年均收益率和总风险值综合评估每次投资的优劣。再次，为了寻找较好的参数组合，分别作出不同参数组合下的收益率等高线和风险等高线，并观察收益率和风险对参数的敏感程度。最后，对参数进行优化，得到最佳参数组合。本文用万科A股票进行实验，结果表明，收益率对P比较敏感，对Q不太敏感，而风险对P和Q都比较敏感；若采用RAROC指标，最佳参数组合为P=2、Q=5；对于不同类型的投资者，根据有效边界和无差异曲线，得到其相应的最佳参数组合，其中，对于保守型投资者来说，最优的参数组合为P=5、Q=4；对于进取型投资者来说，最优的参数组合为P=3、Q=4。 Security technical analysis is a method which is widely used in security investmentanalysis. According to the technical index, investors predict the trend of the stock market inorder to gain more returns, at this time, for the most part, is to find the reversal point of thestock price. If it is earlier for investors to judge price reversal point, and they buy stock at lowprice and sell stock at high price, then they will receive generous returns. The relationshipbetween trading volume and stock price plays an important role in technical analysis. Thispaper designs a model to identify reversal point based on the volume and price. This modelhas a referential significance for investors to make a better investment decision. According to the theory of the technical analysis, and the cointegration relationshipbetween closing price and trading volume, this paper designs a dual index minimuminvestment model considering the closing price and trading volume at the same time. Thismodel is called the bipolar model for short. This bipolar model has two parameters, includingP and Q, where P represents today’s closing price is minimal in P continuous days, Qrepresents today’s trading volume is minimal in Q continuous days. The differentcombination between P and Q represents different investment model. Given P and Q, iftoday’s closing price is minimal in P continuous days and today’s trading volume is minimumin Q continuous days, then the bipolar index value is1, otherwise it is0. When the bipolarindex value changes from1to0, then buy stock. After buying up the stocks, when stock pricedrops from a new high position, and from that position, the stock price drops5percent within4days, then selling all the stocks. When the stock closing price sequence and trading volumesequence are integration of first order, and these sequences exist first order autocorrelationafter first difference, this paper proves the rationality of the bipolar model. Let investment time be n years, using the bipolar model to simulate to invest. First, wecalculate the investment returns at each sale under different P and Q combination, andcalculate the investment returns for n years according to compound interest, then we will getthe annual average rate of return. Second, we calculate daily VaR of each investment based onthe EGARCH model, making the99/%quantile of the VaR series as the total risk value for nyears. Evaluating each investment by the annual average rate of return and total risk value. Third, in order to find better combination of parameters, we make the contour of the returnrate and the risk respectively under different parameters, then observe the sensitivity of theparameters to the return rate and risk. At last, by optimizing these parameters, we obtain thebest combination of parameters. Take Wanke-A stock for a test, the results show that thereturn rate is more sensitive to P and less sensitive to Q, but the risk is sensitive to both P andQ. By using of RAROC index, the best combination of parameters is when P equals2and Qequals5; for different types of investors, according to the efficient borderline and thenon-discrimination curve, we can obtain the corresponding optimal parameter combination.For the conservative investors, the optimal combination of the parameters is when P equals5and Q equals4; for aggressive investors, the optimal combination of the parameters is when Pequals3and Q equals4.

关 键 词： 量价关系 协整分析 模型 优化

分 类 号： [F224 F830.91]

领　　域： [经济管理] [经济管理]