In the previous lessons, we have learned topics from Autocorrelation, Moving Average, and Differencing, to ARIMA Models and Multiple Regression. And we have applied these methods to various datasets, to better understand how to apply this in the real world. However, there is a whole other world in the field of forecasting epidemic diseases outside of ARIMA. So, to summarize and finish the ARIMA part of this course, I would like to talk about the importance of forecasting, what to look forward to in the future, and next steps to learn more.

How to utilize forecasting principles in the real world and How is this relevant

I am sure you all remember the COVID-19 pandemic, and the possible catastrophic effects it had. This is a poor example as it was a new disease and we didn’t have much data on it at the time, so we will get back to that.

But let’s consider an outbreak of the FLU, we already have plenty of data on that, including hospitalizations (when and specifically which locations got impacted), and multiple other statistically measurable metrics have been recorded. What if there was a way for scientists and doctors to be able to not only predict the next FLU outbreak, but also in which location that will occur, over time?

This would be of tremendous interest to governments as it would give them a ‘heads up’ or advance to disease outbreaks, giving them extra time to prepare a response.

This was actually one of the reasons the COVID-19 Pandemic had such wide spread effects, because we weren’t ready to battle such a disease. The nations around the world weren’t prepared for it, which resulted in both a panicked state by the public, and late decisions by government officials, something that if made earlier, could of saved millions.

Ultimately, the field of epidemiology focuses on trying to solve this, or at least in part, minimize it. There are plenty of other methods and theories that have been developed to tackle this issue, some of which we will discuss later.

Next Steps

There are plenty of other topics to discuss here, but, for now, I want to limit this course on topics related to ARIMA models.

I am currently working on a final lesson which will involve the below two topics. So, please stay in tune to hear more about that!

Auto-ARIMA Models

SIR Models and Dynamic Systems