Algorithmic Trading Models, Aversion, and the Indian Scenario

Abstract

Algorithms are automated systems that are designed to do specific tasks. Algorithms are used in a variety of commercial domains due to their efficiency and expanding demand due to the availability of vast amounts of data. Algorithm trading, often known as algo trading, is a significant application in finance. There has been very little research on the many sorts of algorithms utilized for specialized stock trading purposes. This exploratory study examines the literature over the last five years (2019-2023) on various models of algorithms used in stock trading. The study discovered that algorithm trading models were applied mostly to predicting and, to a lesser extent, classification. The most used models are long- Short-Term Memory (LSTM), Support Vector Machines (SVM), Gradient Boosting (GB), and Recurrent Neural Networks (RNN). Furthermore, algorithms have been discovered to improve the accuracy of portfolio strategies. The paper then delves into the causes of algorithm aversion and its dimensions of stock trading. The Indian environment appears to be shifting in favor of algo trading, with regulatory action to protect investors and the stock market's interests.

Keywords: algorithm trading, algo trading, algorithm aversion, algorithm in finance.

JEL Classification: G10, C11