Smart trading : leveraging machine learnings to improve profitability of technical trading rules in the Thailand SET100 index

Name: Jantakarn Wannasuk

keyword: Stock trading strategies

; Technical analysis

; Random forest

; Artificial neural networks

; Technical indicators

Abstract: The Thai stock market, represented by the SET100 index, has recently faced a downturn, prompting traders to seek innovative strategies to navigate these challenging conditions. This study explores the potential of machine learning-enhanced trading strategies to improve profitability, with a particular focus on sector-specific approaches. We analyzed a comprehensive dataset of 41 high-liquidity, large-cap SET100 stocks from January 2016 to December 2023. To identify optimal entry and exit points for stock transactions, we integrate technical indicators, specifically the Relative Strength Index (RSI) and Moving Average Convergence Divergence (MACD), to generate trading signals. These signals are then analyzed using Random Forest (RF) and Artificial Neural Network (ANN) machine learning models. Our analysis reveals that the effectiveness of machine learning integration with technical indicators varies significantly across different market sectors. In the FOOD and HELTH sectors, the RF-RSI strategy yielded substantial return improvements compared to using RSI alone. Additionally, the ANN-RSI strategy demonstrated the ability to potentially avoid losses in the TRANS sector compared to relying on RSI alone. Integrating machine learning with MACD produced mixed results. Both ANN-MACD and RF-MACD strategies reduced losses in the HELTH sector, suggesting potential benefits in mitigating downward trends. However, in some sectors, using MACD alone might be sufficient based on the specific dataset used. This highlights the need to consider the historical performance of technical indicators within each sector when determining if machine learning integration can offer additional value. Our research offers valuable insights for investors seeking to optimize profits in challenging market conditions. The findings demonstrate the potential for sector-specific, machine learning-enhanced trading strategies to improve profitability. Investors can leverage these insights by carefully selecting machine learning models and technical indicators that align with the characteristics of their targeted sectors.

Thammasat University. Thammasat University Library

Address: BANGKOK

Email: preserv@tu.ac.th

Name: Rujira Chaysiri

Created: 2024

Modified: 2024-11-26

Issued: 2024-11-26

วิทยานิพนธ์/Thesis

application/pdf

DOI: 10.14457/TU.the.2023.545

eng

DegreeName: Master of Science

Level: Master's Degree

Descipline: Engineering and Technology

Grantor: Thammasat University

©copyrights Thammasat University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	15704Jantakarn.pdf	1.31 MB	1	2025-05-17 17:44:11

ใช้เวลา

0.027266 วินาที