📈 Asset Price Prediction using Machine Learning

This project aims to predict the future price of a stock (Google) using a combination of traditional and modern machine learning approaches including LSTM (deep learning) and XGBoost.

🚀 Objective

To build and compare predictive models for stock price forecasting using:

• Time Series Analysis
• Supervised Learning Models
• Deep Learning Architectures (LSTM)

📊 Dataset

• Source: Google Stock Price - Kaggle Dataset
• Description: Historical stock prices of Google including Open, High, Low, Close, and Volume.

🛠️ Tools & Technologies

• Python
• NumPy, Pandas, Matplotlib
• Scikit-learn
• XGBoost
• TensorFlow / Keras
• ARIMA

🔍 Project Workflow

1. Data Preprocessing

   • Handling missing values
   • Feature scaling
   • Train-test split

2. Exploratory Data Analysis

   • Visual trends in closing price
   • Moving average smoothing

3. Model Implementation

   • 📦 XGBoost Regression
   • 🔁 LSTM Neural Network
   • 📉 ARIMA Model for traditional forecasting

4. Model Evaluation

   • Metrics: MAE, RMSE
   • Visualization of predicted vs actual stock prices

📈 Results

• LSTM shows strong temporal learning ability and performs better on longer-term sequences.
• XGBoost captures short-term trends effectively.
• RMSE and MAE were used to compare models.

✅ Deliverables Covered

• ✅ Time Series Forecasting
• ✅ Financial Data Handling
• ✅ Predictive Modeling (ARIMA, XGBoost, LSTM)
• ✅ Model Evaluation & Visualization

📚 References

• YouTube: LSTM for Stock Prediction
• YouTube: ML-based Stock Prediction

---

📌 Note: This project is a learning exercise in time series forecasting and model comparison. Predictions should not be used for actual financial decisions.