One of the most notable "deep" papers in the technical domain is "Deep Stock Trading: A Hierarchical Reinforcement Learning Framework for Portfolio Optimization and Order Execution", which uses complex AI policies to decide both what to trade (the high-level portfolio) and how to execute the buy/sell orders (the low-level timing). 1. AI and Deep Learning Models
Research explores how external "noise" dictates trading signals.
Research from the American Economic Association shows that investors are generally more likely to sell a security when they have a gain rather than a loss, even if holding onto the "loser" is financially detrimental.
The HRPM Framework (mentioned above) is significant for managing the "commission fee" problem—it learns that simply knowing whether to buy or sell isn't enough; you must also minimize the cost of the trade itself. 2. Behavioral & Decision-Making Research
Studies have found that combining Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) models offers higher accuracy than simple time-series methods.
One of the most notable "deep" papers in the technical domain is "Deep Stock Trading: A Hierarchical Reinforcement Learning Framework for Portfolio Optimization and Order Execution", which uses complex AI policies to decide both what to trade (the high-level portfolio) and how to execute the buy/sell orders (the low-level timing). 1. AI and Deep Learning Models
Research explores how external "noise" dictates trading signals. sell or buy
Research from the American Economic Association shows that investors are generally more likely to sell a security when they have a gain rather than a loss, even if holding onto the "loser" is financially detrimental. One of the most notable "deep" papers in
The HRPM Framework (mentioned above) is significant for managing the "commission fee" problem—it learns that simply knowing whether to buy or sell isn't enough; you must also minimize the cost of the trade itself. 2. Behavioral & Decision-Making Research Research from the American Economic Association shows that
Studies have found that combining Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) models offers higher accuracy than simple time-series methods.