← Publications

A Hybrid Method based on Machine Learning to Predict the Stock Prices in Bangladesh

2022 IEEE 13th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON)

Md. Nabil Rahman Khan, S.M. Raiyan Reza, , Most. Sadia Salsabil, Khan Md. Hasib, Mohammad Shafiul Alam

IEEE IEMCON 2022Published

Table of contents

Abstract

We predict next-day closing prices on the Dhaka Stock Exchange with a hybrid model. A stacked LSTM forecasts the price, a GARCH(9,13) model forecasts volatility, and a linear regression combines both with moving averages and daily returns. Trained on over a million rows of DSE data from 2007 to 2020, the hybrid beat linear regression and the stacked LSTM on 10 of the 11 stocks we tested.

Hybrid model architecture: moving averages, a stacked LSTM and GARCH volatility feeding a linear regression
Hybrid model architecture: moving averages, a stacked LSTM and GARCH volatility feeding a linear regression.

Key findings

  • Lowest RMSE on 10 of 11 stocks, against linear regression and a stacked LSTM on their own
  • Mean RMSE of 0.49, compared with 0.95 for linear regression and 4.34 for the stacked LSTM
  • GARCH order (9,13) picked by the lowest AIC across every (p, q) pair from 1 to 15

Results

AB Bank
AMCL (PRAN)
Apex Tannery
BD Finance
Grameenphone
Padma Oil
Popular Life
Samorita
RAK Ceramics
GBB Power
Beximco
RMSE on the test set for each stock. Lower (further left) is better. Tap a model to hide it; the axis rescales.
RMSE on the test set for each stock (lower is better).
StockLinear RegressionStacked LSTMHybrid
AB Bank0.571.480.12
AMCL (PRAN)1.525.060.90
Apex Tannery0.956.380.50
BD Finance0.541.050.24
Grameenphone2.3011.681.40
Padma Oil1.286.340.74
Popular Life0.845.190.37
Samorita1.052.760.35
RAK Ceramics0.683.090.27
GBB Power0.521.300.11
Beximco0.223.360.42

Citation

Published in 2022 IEEE 13th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON). DOI: 10.1109/IEMCON56893.2022.9946610

View on IEEE XploreRead the blog post
  • Machine Learning
  • LSTM
  • GARCH
  • Time-Series Forecasting
  • Financial ML

Co-authored by

Software Engineer in Dhaka, Bangladesh. Skilled in AI agents, automation bots, and OCR solutions for intelligent process automation.

View profile

Comments (0)

You’ll sign in when you post. Only your name and photo are shown.

Comments are checked by AI before they appear. Read the rules
  • Be respectful: no insults, harassment, hate or threats. Disagreeing is fine.
  • No spam: no ads, self-promotion, SEO or affiliate links, crypto or gambling.
  • Nothing sexual, violent or illegal, and no scams or malware.
  • Don't share anyone's personal details (phone numbers, addresses, private emails).
  • Keep it a real comment: no gibberish, test text or messages aimed at the AI.

Loading comments…