A Contratian Strategy Based On The Predictability Of High And Low Prices Of Securities

In academic research, there is an argument about whether technical analysis is valid. In recent years, more and more studies of the technical analysis provide support evidences by theory and models. Technical researchers mostly use the closing price of the securities to analyze, however, several studies have shown that a combination of the highest price and lowest price could raise the predicting method, which is more meaningful than simply based on the closing price. Moreover, technical indicators such as K-line, KD indicators, pressure support line are also closely related with the highest and lowest prices. Therefore, this article will focus on research of a trading strategy based on the predicted highest price and the lowest price.At present, most of the domestic researches on cointegration of time series data are limited to integer-dimensional differential framework, which will result in inaccurate analysis results. In this article, we find out the fractional cointegration relationship between the highest and lowest prices of securities, and apply a fractional cointegrating model to the modeling and forecasting of the highest and lowest prices of securities. Then make empirical studies which show that the fractional cointegration model will raise the ability to help investors get higher return and lower risk in transaction.In this paper, I choose the daily data of IF300 index futures during five years as the data sample to verify the fractional cointegration relationship between the highest price, lowest price, and the long memory feature of range, indicating the high and low prices are predictable. Then use the fractional cointegration vector error correction model (FVECM) to forecast, and build an intraday reversal strategy based on the principle of support and pressure lines. I use the 1 minute sharing price data of IF300 stock index futures to simulate the reversal strategy, prove that FVECM predictions will get excess returns compared to other models, providing a new way to the quantitative investment on the transaction in security market.The main conclusions of this paper is that the prediction of the highest price and lowest price of securities can increase revenue in transactions and reduce the risk in transaction. However, there are also shortcomings and limitations of this paper. I hope that in future researches, we can see more flexible model that can achieve the prediction of the highest price and lowest price, and apply to pricing some other derivations. In addition, since the strategy in this article is based on intraday trading and Chinese stock market is limited by the T+1 system, we can’t expand this strategy to the stock market. I believe that an improvement based on this paper will make the strategy in this article more practical.