The CTA Quantitative Investment Model Based On DTW Algorithm

This paper combines the dynamic time warping algorithm with the technical analysis theory and applies it to the forecast of futures daily return rate,thus constructing a quantitative investment strategy model and optimizing the model so that the model before and after optimization is better than the market.Good performance,compared with the annual decline of 16.09% in the market,the average profit of 6% or more before and after optimization,and the evaluation indicators are better than the broader market,which shows the effectiveness of the model.Artificial intelligence technology is becoming more and more mature,gradually entering people's lives,and becoming an indispensable part of human beings.The most common is the speech recognition system on mobile phones.As the core algorithm of the system,the essence of the dynamic time warping algorithm is to measure the similarity between two data sequences.In today's data age,it is to measure the similarity between two time series,which is represented by two on the coordinate axis.The degree of similarity of the waveform.Whether it is time series data or waveforms,it is a very common thing for the secondary market.Each line on the K line can generate large amounts of data and waveforms.Therefore,if the algorithm is applied to the K line.On,it seems to be feasible.Combined with the traditional theoretical basis of technical analysis,whether it is foreign Dow Theory,wave theory,K-line theory and trend theory,or domestic entanglement and periodic theory,it is related to waveform,and this waveform is also derived from K.The line sequence,at the same time,all of these theories have a hypothesis that history will repeat.In this way,the DTW algorithm is applied to the K line,and there is enough theoretical support.The original intention of this model is to match the current waveform with the historical waveform,which is the historically repeated hypothetical data restoration.It is.When it comes to the K-line,for the domestic market,although both the stock market and the futures market can be invested,the futures market has more advantages than the stock market,so that the investment target of this article is transferred to the futures.At the same time,the evasive effect of risk also makes the emergence of portfolio investment ideas in the model.Although there are still many shortcomings in the structure and content of this paper,as the first case of applying the dynamic time warping algorithm,that is,DTW to the development of CTA strategy in the secondary market,the merits are also there.The development of strategy plays a good role in learning.