And the last task that we will consider is the prediction of the EEG signal. Time series in medicine are very important data, the analysis of which can solve many problems. One of them is the analysis of the EEG signal for the prediction of epileptic seizures. The example we'll look at uses the ARIMA-GARCH predictive model.

We will not delve into the operation of this model in detail but only mention that the GARCH model can capture the flat period and fluctuation period of time series.

Processing medical temporal data is no different from a normal data processing algorithm: you still need to remove noise, fill in missing data, and turn non-stationary data into stationary data.


This small section was devoted to a brief overview of various real-world problems for which, as you understand, models such as moving average, autoregressive model, and ARIMA are just a starter. In the future, these models will be modified depending on the type of problem, and sometimes they are not at all effective enough to solve certain problems.

Is it possible to solve the problem of predicting the EEG signal using an approach similar to the problem with predicting the behavior of clients (instead of clients - individual records of patients)?

What can be done with thousands of online store purchase records? How can we analyze this data and predict its growth? In this course, you will learn what parameters we can analyze in time series and how to create predictive models. Let's get started!

Start from the introduction to time series to move on to more in-depth topics. Find out what it is, how often time series occur in real business, how to analyze them and make forecasts.

What is the difference between trend and seasonality? Is the data I'm working with stationary or not? In this section, you will learn how to analyze characteristics of time series using Python!


Learn more about time series through… their visualization! Explore your data with 2D and 3D graphs.


What will be our first step? Let's start with stationary models! Create your first predictive model that will forecast time series.

Most of the data you work with is non-stationary. Learn how to process such data and what models to use to predict it.

Let's dive into a variety of interesting tasks that you may encounter while working with time series! Find out how big these challenges are and what models we can use to solve them.

Time Series Analysis

Medicine: EEG Forecasting

Time Series Analysis

Medicine: EEG Forecasting