Project Showcase
Highlight of my featured works
🥑📈 Avocado Price Forecasting
A time series forecasting project that analyzes avocado price trends and predicts future prices using statistical methods and deep learning models.
🎯 What I Learned
- Performed seasonal decomposition and stationarity tests
- Applied RobustScaler to stabilize training
- Implemented neural network forecasting with hyperparameter tuning
🛠 Tech Stack
- Python, Pandas, NumPy
- Matplotlib, Seaborn, Statsmodels
- Scikit-learn
- TensorFlow / Keras, Keras Tuner
📌 Evaluation / Next Step
The tuned Keras model achieved lower MSE/MAE and higher R² compared to the untuned version, thanks to scaling and hyperparameter optimization. However, sudden market shifts remain challenging. Future work includes experimenting with LSTM/GRU architectures and hybrid models that combine ARIMA with neural networks.