Deep Stock Representation Learning: From Candlestick Charts to Investment Decisions

Abstract

We propose a novel investment decision strategy (IDS) based on deep learning. The performance of many IDSs is affected by stock similarity. Most existing stock similarity measurements have the problems: (a) The linear nature of many measurements cannot capture nonlinear stock dynamics; (b) The estimation of many similarity metrics (e.g. covariance) needs very long period historic data (e.g. 3K days) which cannot represent current market effectively; (c) They cannot capture translation-invariance. To solve these problems, we apply Convolutional AutoEncoder to learn a stock representation, based on which we propose a novel portfolio construction strategy by: (i) using the deeply learned representation and modularity optimisation to cluster stocks and identify diverse sectors, (ii) picking stocks within each cluster according to their Sharpe ratio (Sharpe 1994). Overall this strategy provides low-risk high-return portfolios. We use the Financial Times Stock Exchange 100 Index (FTSE 100) data for evaluation. Results show our portfolio outperforms FTSE 100 index and many well known funds in terms of total return in 2000 trading days.

DPP: Deep predictor for price movement from candlestick charts

“…Reading candlestick charts allows traders to understand impacts of trends called visual features rather than reading numerical raw data directly. Some recent investigations manifest that analyzing visual characteristics of candlestick charts to predict stock market is still a hot topic, for examples, Hu et al [8] summarize the historical financial data as images using candlestick charts and adopt the convolutional autoencoder for feature learning from the image data, Fengqian and Chao [25] apply K-line theory to characterize candlesticks as a generalization of price movements over a time period and then propose the deep reinforcement learning based system to reach adaptive control in the unknown environment, and Ananthi and Vijayakumar [26] proposed a system that generates signals on the candlesticks to predict market price movement by using regression and candlestick pattern detection. Furthermore, a novel variation of conventional candlesticks, RGBSticks, is introduced in [27] to predict the daily stock price of a company by using an autoencoder based on deep neural network.…”

“…It is easy to recognize visual trends from candlestick charts and use the corresponding information collected to help predict price movements. Although the literature has shown that candlestick pattern analysis is a successful approach in financial forecasting [7][8][9][10], predicting the price movements by reading the visual trends from candlestick charts is still challenging. One can argue that nuances are lost in the candlestick charts while candlestick charts simplifies complex analysis behind the scene.…”

YOLO Object Recognition Algorithm and “Buy-Sell Decision” Model Over 2D Candlestick Charts

“…Huy et al (2017) developed a new prediction model based on both online financial news and past stock price data to predict stock movements in advance [8]. Hu et al (2017) proposed a novel investment decision strategy based on deep learning. Key idea is to endow an algorithmic strategy with the ability to make decisions with a similar kind of visual cues used by human traders.…”

“…Key idea is to endow an algorithmic strategy with the ability to make decisions with a similar kind of visual cues used by human traders. To this end we apply Convolutional Auto Encoder (CAE) to learn an asset representation based on visual inspection of the asset's trading history [18].…”

Fundamental factors of firm value: Evidence from the cement industry in Indonesia

“…The Investors who invest in the stock market will analyze the condition of the company to get maximum profit (Hu et al, 2018;Lee, Cho, Kwon, & Sohn, 2019). The Investors will look for prospective companies by considering external and internal factors (Rao, Chandy, & Prabhu, 2008;Sujoko, 2017).…”