JoaoESmoreira

Time Series Analysis and Forecasting

This project focuses on the time series analysis of mean temperature data for a selected location. The goal is to model and forecast temperature trends using different time series analysis techniques, including exponential smoothing and other machine-learning methods, to assess their predictive accuracy.

In addition to forecasting mean temperature, we extend the analysis by incorporating multivariate models that consider other weather variables. This approach allows for a more comprehensive understanding of the relationships between different weather factors, ultimately leading to more reliable forecasts.

The full repository and documentation can be found here: Time Series Analysis Project Repository.

Tools and Technologies

To run the project correctly, the following technologies are required:

python 3.10+
scikit‑learn
pandas
numpy
matplotlib
seaborn

Data

The dataset used for this project was obtained from the Open-Meteo Historical Weather API, which provides daily weather observations for various locations worldwide. For this analysis, we selected data from Tokyo for the period spanning from January 1st, 1940, to the present. The primary variable of interest is the mean daily temperature, supplemented with additional weather variables for multivariate analysis.

The dataset contains daily observations structured with a date column and corresponding values for each weather variable.

Repository Structure

The repository is organized into several main directories to maintain clarity and separation of concerns throughout the project:

src/ – Contains all the project’s source code, including scripts and Jupyter notebooks. This is where the main logic and implementation of the project are developed.
data/ – Stores all datasets and input files used in the experiments.
docs/ – Holds documentation/report files
delevery/ – Includes final deliverables such as reports, slides, and results ready for submission or presentation.
requirements.txt – Lists all dependencies required to run the project in a reproducible environment.

This structure ensures that the project remains modular, easy to navigate, and reproducible across different environments.

License

This project is distributed under the [MIT](LICENSE) License.

Footer

The contents of this website are licensed under the Creative Commons Attribution-NoDerivatives 4.0 International License (CC-BY-ND 4.0).

The source code of this website is licensed under the MIT license, and available in GitHub repositor. User-submitted contributions to the site are welcome, as long as the contributor agrees to license their submission with the CC-BY-ND 4.0 license.