Introduction

The Earth's magnetic field plays a crucial role in various aspects of our planet, acting like a giant bar magnet with a magnetic north and south pole. This magnetic field undergoes changes in strength and orientation over time, a phenomenon referred to as "geomagnetic secular variation." In recent years, the geomagnetic poles have been drifting from Canada towards Siberia at an accelerating rate, possibly influenced by a strong positive geomagnetic anomaly near Lake Baikal in Siberia. This drift presents significant challenges to global navigation systems and various industries. Moreover, changes in the magnetic field intensity at higher altitudes have a direct impact on satellite performance and all dependent equipment. The magnetic field also protects Earth's inhabitants from solar flares, magnetic storms, and cosmic rays, highlighting the importance of studying its behavior. In this project, our primary goal is to develop a forecasting methodology based on the Box-Jenkins framework to predict the magnetic intensities of the North and South poles at their current locations.

Project Objectives and Learning Outcomes

This project aims to provide you with hands-on experience in dealing with the challenges faced by a time series analyst in real-world situations. Through this project, you will have the opportunity to hone your skills in independent thinking, decision-making, and validating the outcomes of your analysis without relying on external sources for verification. While your instructor will be available during office hours to offer guidance, the emphasis is on developing your abilities as an independent researcher, thinker, and problem solver in the field of time series analysis.

Written Report Structure and Guidelines

Your report should be well-organized and include the following sections:

•    Jupyter Notebook: Create your report using Jupyter Notebook and submit a PDF or HTML version on Crowdmark. Organize your report into distinct sections and use various headings and typesetting to enhance your work's presentation.

•   Title and Abstract: Select an appropriate title and provide a concise abstract summarizing the project's purpose, main points, and significant findings (maximum 200 words).

•   Introduction: Introduce the dataset and its relevance, explaining why investigating this problem is essential.  Provide  pertinent  background  information  and  formulate  critical  questions  to  address throughout your analysis (maximum 1,000 words).

•   Model Specification: Clearly describe the data analysis process used to specify your candidate models, guiding the reader through your thought process that leads to your model selections.

•   Fitting and Diagnostics: Detail the model fitting and diagnostics techniques employed, aiming to identify a "final" model for forecasting. Discuss any potential deficiencies or limitations in your final model.

•   Forecasting: Describe the techniques used to forecast future observations, explaining the importance of forecasting and the potential impacts and implications of your forecasts.

•   Discussion:  Summarize your project, drawing main conclusions and discussing any shortcomings, limitations, or challenges encountered during your analysis.

•   Bibliography: Accurately cite all references used throughout your report.

•   Appendices: Include supplementary graphics, plots, or output that may be informative to the reader but would interrupt the flow of the main body of the report.