Stock Price Forecasting using ARIMA in Python

This repository demonstrates how to forecast stock prices using the ARIMA (AutoRegressive Integrated Moving Average) model in Python. The stock market data of HCL from April 2017 to April 2022 is used as a case study to predict the future stock prices.

Table of Contents

• Introduction
• Data Information
• Dependencies
• Installation
• Usage
• Results
• Further Reading
• Contributing
• License

Introduction

The stock market is highly volatile and unpredictable, making it challenging to forecast stock prices accurately. In this project, we use the ARIMA model, a class of statistical models for analyzing and forecasting time series data, to predict future stock prices.

Data Information

The dataset used in this project consists of stock market data for HCL from 2017-04-21 to 2022-04-21. The dataset includes various attributes such as Open, High, Low, Close, Adj Close, and Volume. The 'Adj Close' column is used for predicting the future stock prices.

Dependencies

• Python 3.x
• pandas
• numpy
• matplotlib
• statsmodels
• pmdarima

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/StockPriceForecasting_ARIMA.git

Certainly! Below is a suggested README file for your GitHub repository on forecasting stock prices using ARIMA in Python. I've also included some potential names for your repository.

Repository Name Suggestions:

1. StockPriceForecasting_ARIMA
2. ARIMA_StockPrediction
3. StockPricePredictor
4. StockForecasting
5. ARIMA_StockAnalysis

README.md

# Stock Price Forecasting using ARIMA in Python

This repository demonstrates how to forecast stock prices using the ARIMA (AutoRegressive Integrated Moving Average) model in Python. The stock market data of HCL from April 2017 to April 2022 is used as a case study to predict the future stock prices.

## Table of Contents
- [Introduction](#introduction)
- [Data Information](#data-information)
- [Dependencies](#dependencies)
- [Installation](#installation)
- [Usage](#usage)
- [Results](#results)
- [Further Reading](#further-reading)
- [Contributing](#contributing)
- [License](#license)

## Introduction
The stock market is highly volatile and unpredictable, making it challenging to forecast stock prices accurately. In this project, we use the ARIMA model, a class of statistical models for analyzing and forecasting time series data, to predict future stock prices.

## Data Information
The dataset used in this project consists of stock market data for HCL from 2017-04-21 to 2022-04-21. The dataset includes various attributes such as Open, High, Low, Close, Adj Close, and Volume. The 'Adj Close' column is used for predicting the future stock prices.

## Dependencies
- Python 3.x
- pandas
- numpy
- matplotlib
- statsmodels
- pmdarima

## Installation
1. Clone the repository:
   ```bash
   git clone https://github.com/yourusername/StockPriceForecasting_ARIMA.git

2. Navigate to the project directory:

   cd StockPriceForecasting_ARIMA

3. Install the required dependencies:

   pip install -r requirements.txt

Usage

1. Load the dataset:

   import pandas as pd
   
   data = pd.read_csv('path/to/your/data.csv')

2. Check for missing values:

   data.isnull().sum()

3. Visualize the data:

   import matplotlib.pyplot as plt
   
   plt.plot(data['Adj Close'])
   plt.title('Stock Prices of HCL')
   plt.show()

4. Check for stationarity using Augmented Dicky Fuller (ADF) test and make the data stationary if needed:

   from statsmodels.tsa.stattools import adfuller
   
   result = adfuller(data['Adj Close'])
   print('ADF Statistic:', result[0])
   print('p-value:', result[1])

5. Apply the ARIMA model:

   from pmdarima import auto_arima
   
   model = auto_arima(data['Adj Close'], seasonal=False)
   model.summary()

6. Make predictions and evaluate the model:

   train = data[:int(0.75*len(data))]
   test = data[int(0.75*len(data)):]
   
   model.fit(train['Adj Close'])
   predictions = model.predict(n_periods=len(test))
   
   plt.plot(test['Adj Close'].values, label='Actual')
   plt.plot(predictions, label='Predicted')
   plt.legend()
   plt.show()

Results

The ARIMA model with optimal parameters (p=0, d=1, q=1) is used to predict the stock prices. The model performance is evaluated using metrics such as Mean Squared Error (MSE), Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE).

Further Reading

• Autoregressive Integrated Moving Average - Wikipedia
• What is ARIMA and SARIMA Model?
• Autocorrelation and Partial Autocorrelation in Time Series Data

Contributing

Contributions are welcome! Please open an issue or submit a pull request for any improvements or suggestions.

License

This project is licensed under the MIT License - see the LICENSE file for details.