Development of Stock Recommendation Method

In this paper, we propose a method for recommending appropriate combinations of stocks based on the waveforms of stock price changes. To invest in stocks while reducing risk, specialized knowledge is required. To solve the problem, a method for stock recommendation with different characteristics is desired. Stock price movements can be captured as a waveform. Dynamic Time Warping (DTW), cross-correlation functions, and fast Fourier transforms are used to compare the features of the waveforms. Combining stocks with different waveforms avoids the risk of simultaneous crashes.