Skip to content

Multivariate Time Series ARIMA and LSTM Models Forecasting Indonesian Bank Stocks (BBCA, BBRI, BMRI)

License

Notifications You must be signed in to change notification settings

TnMarty2/Multivariate-Time-Series-DeepLearning

Repository files navigation

Multivariate-Time-Series-DeepLearning

Multivariate Time Series ARIMA and LSTM Models Forecasting Indonesian Bank Stocks (BBCA, BBRI, BMRI)

📌 Project Overview

This repository aims to compare traditional time series forecasting (ARIMA) with deep learning models (LSTM variants) for multivariate stock price prediction. The dataset includes stock prices and trading volumes of three major Indonesian banks:

  • BBCA (Bank Central Asia)
  • BBRI (Bank Rakyat Indonesia)
  • BMRI (Bank Mandiri)

The goal is to evaluate forecasting performance and identify which model provides the most accurate predictions for financial time series data.


🧠 Models Implemented

  • ARIMA (AutoRegressive Integrated Moving Average)
  • Vanilla LSTM
  • Bidirectional LSTM
  • Stacked LSTM

Each model was trained on preprocessed multivariate data (Close, Volume, etc.) and evaluated based on prediction accuracy.


🔧 Features & Workflow

1. Data Collection

  • Source: Kaggle
  • Timeframe: Historical daily prices

2. Data Preprocessing

  • Handling missing values
  • Normalization (MinMaxScaler)
  • Feature selection: Close, Volume
  • Sequence generation for LSTM models

3. Modeling

  • ARIMA for univariate time series
  • LSTM variants for multivariate time series
  • Train-test split ratio 80:20

4. Evaluation

  • Root Mean Squared Error (RMSE)
  • Mean Absolute Error (MAE)
  • Mean Absolute Percentage Error (MAPE)

5. Visualization

  • Time series plots
  • Prediction vs. actual comparison
  • Loss and metric trend lines

📚 Libraries Used

  • Data Handling: pandas, numpy
  • Visualization: matplotlib, seaborn
  • Preprocessing & Metrics: scikit-learn
  • Time Series Modeling: statsmodels
  • Deep Learning: tensorflow, keras
  • Data Source: yfinance

🚀 Results Summary

BBCA Close

Model RMSE MAE MAPE
ARIMA 124,1 95,0 0,9%
Vanilla LSTM 200,5 160,7 1,6%
Bidirectional LSTM 227,8 192,0 1,9%
Stacked LSTM 231,5 188,0 1,9%

BBCA Volume

Model RMSE MAE MAPE
ARIMA 52.662.514,0 27.808.104,2 36,9%
Vanilla LSTM 55.683.352,3 26.871.570,6 30,4%
Bidirectional LSTM 53.678.671,9 26.862.275,7 34,4%
Stacked LSTM 54.245.496,5 28.869.947,1 40,7%

About

Multivariate Time Series ARIMA and LSTM Models Forecasting Indonesian Bank Stocks (BBCA, BBRI, BMRI)

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published