Predicting Stock Market Volatility from Candlestick Charts: A Multiple Attention Mechanism Graph Neural Network Approach

Table 1

Analytical methods for processing graphical indicators of securities market.

Literature	Research content	Experiment		Focus
		Features	Target	Single-day candlestick features	Time series correlation
[25]	An analysis of candlestick chart composition and characteristics was conducted, focusing on visual representations and financial technical analysis tools.	High, low, open, and close prices	Visual analysis	Yes	No
[26]	Developed an expert system for candlestick chart analysis, or chart interpreter, to predict the best market timing. This expert system has patterns and rules for predicting future stock price movements. Depending on their meaning, the defined patterns can be divided into five groups: Down patterns, up patterns, neutral patterns, trend continuation patterns, and trend reversal patterns.	Predefined patterns	Predicting stock price movements	No	No
[27]	Based on the color, size, and relative position of the candlesticks, these features are combined into a tree trading strategy using the Chi-square automatic interaction detector (CAD) algorithm. The features and methods constructed for candlesticks proved to be effective in identifying the candlestick patterns.	Color, size, and relative position of candlesticks	Predicting buy, sell, or hold	No	Yes
[28]	Two models were constructed. Model 1 is a committee machine with a generalized regression neural network (GRNN). Model 2, on the other hand, is a hybrid fuzzy logic-controlled network for identifying candlestick charts and predicting stock market conditions.	Color, size, and relative position of candlesticks	Predicting stock price movements	No	Yes
[29]	Extracted the sentiment features of social media, constructed historical time series data candlestick charts, integrated candlestick charts and social media data, and proposed a multichannel collaborative network based on convolutional neural networks for stock price trend analysis.	Candlestick charts	Predicting stock price movements	No	No
Ours	Embedding stock market candlestick chart graphs to fully represent graphical indicator features, setting a single-day candlestick chart as a node. Edges are created between adjacent single-day candlestick chart. Multiple attention graph neural networks are introduced for stock market price volatility prediction based on the constructed stock market graphical indicator candlestick chart data.	Color, upper shadow, lower shadow, and solid length	Predicting stock price movements	Yes	Yes