Skip to content

Latest commit

 

History

History
21 lines (18 loc) · 1.56 KB

File metadata and controls

21 lines (18 loc) · 1.56 KB

Time Series Analysis and Forecasting with LSTM

Objective

The goal of this notebook is to predict future values for each column in the dataset using Long Short-Term Memory (LSTM) models.

dataset

this is my first time working with timeseries, so I appreciate any comments and suggestions

Outline

In this notebook, I will walk through the following steps:
  • Data Exploration and Forecast Horizon: Initial examination of the dataset, determining the time period for predictions.
  • Detrending: Removing long-term trends from the data.
  • Deseasoning: Eliminating seasonal patterns to focus on the underlying data correlations.
  • Post-Processing Analysis: Examining the data after detrending and deseasoning to ensure these processes were effective.
  • Model Training and Evaluation: Training the LSTM model on the processed data and evaluating its performance.
  • Predicting on Test Data: Making predictions on the test set to check the performance.
  • Model Training on Full Data: Retraining the model on the entire dataset to improve its forecasting ability.
  • Final Prediction: Generating final predictions for future data points using the trained model.
  • Conclusion: Summarizing the findings, discussing the results, and suggesting potential improvements.