DataSynthis_ML_JobTask
This repository contains trained models for time series forecasting of stock prices using ARIMA and LSTM.
Model Details
- Model Types: ARIMA (AutoRegressive Integrated Moving Average) and LSTM (Long Short-Term Memory)
- ARIMA Order: (5, 1, 0)
- Training Data: Historical closing prices of AAL stock from the
all_stocks_5yr.csv dataset, specifically the training split (80% of the data).
- Training Date Range: 2013-02-08 to 2017-02-07
Performance Comparison (on test set)
| Model | RMSE | MAPE |
|---|
| ARIMA | 0.892 | 1.357% |
| LSTM | 8.730 | 0.000% |
How to Load and Use the Models
You can load the serialized models using the pickle library and use them for forecasting.
README.md: This model card providing details about the model.
