This project analyzes and forecasts the stock prices of Apple Inc. (AAPL) and Honeywell International Inc. (HON) using advanced time series forecasting techniques. The analysis spans 252 market days, employing methods like exponential smoothing, weighted moving averages, and regression analysis to predict stock prices and evaluate forecasting accuracy.

1. Short-Term Forecasting:

   • Apply exponential smoothing with varying parameters to predict stock prices.
   • Identify the optimal smoothing constants using error metrics like MAPD and MAPE.

2. Long-Term Forecasting:

   • Use weighted moving averages and linear trend analysis to predict stock prices over extended periods.
   • Evaluate accuracy by comparing forecasted values with actual prices.

3. Regression Analysis:

   • Perform linear regression to model stock prices against time periods.
   • Conduct residual analysis to validate the regression model.

4. Portfolio Optimization:

   • Determine an optimal allocation strategy between AAPL and HON stocks for a balanced portfolio.

• Primary Tool: Microsoft Excel
• Techniques:
  
  • Exponential Smoothing (α and β parameter tuning)
  • Weighted Moving Averages (3-period WMA)
  • Regression Analysis (Simple Linear Regression)
• Metrics:
  
  • Mean Absolute Percentage Deviation (MAPD)
  • Mean Absolute Percentage Error (MAPE)

• Apple Inc. (AAPL):
  
  • Best smoothing constant: α = 0.75
  • MAPD: 1.96%
• Honeywell International Inc. (HON):
  
  • Best smoothing constant: α = 0.75
  • MAPD: 1.78%

• AAPL MAPE: 3.76%
• HON MAPE: 2.69%

• AAPL Regression Model:
  
  • R²: 0.7616, Slope: 0.2437, MAPE: 10.56%
• HON Regression Model:
  
  • R²: 0.0177, Slope: -0.0314, MAPE: 9.76%

• Optimal Allocation:
  
  • AAPL: 60%
  • HON: 40%

. ├── Data/ │ ├── Time_Series_Forecasting.xlsx ├── Analysis/ │ ├── Exponential_Smoothing.xlsx │ ├── Weighted_Moving_Averages.xlsx │ ├── Regression_Analysis.xlsx ├── Visualizations/ │ ├── AAPL_vs_HON_Stock_Prices.png ├── Reports/ │ ├── Time_Series_Forecasting_Report.pdf ├── README.md

For a detailed analysis, including methodology, formulas, and visualizations, refer to the full report:
📄 Time Series Forecasting Report

Feel free to reach out for feedback, questions, or collaboration opportunities:
LinkedIn: Dr. Syed Faizan

Author: Syed Faizan
Master’s Student in Data Analytics and Machine Learning