Time series are used in a multitude of areas, and it is becoming increasingly common for the generation of data to be accompanied by a timestamp. The applications of this type of analysis are innumerable: from the predictive analysis of the evolution of a patient’s vital signs to the prediction of the evolution of a stock market value.

We will begin this course with an introduction to the most basic techniques on which this type of methodologies are based and then we will delve into the ARIMA family of algorithms.

In order to be able to take full advantage of the course, participants should know the basics of the Python programming language.

Course contents:

• Introduction to time series analysis
• Basic time series forecasting methods
• Data loading and transformation
• Mathematical foundation
  • Correlation
  • Autocorrelation
  • Autocorrelation function
  • Partial autocorrelation function
  • White noise
• Auto-regressive (AR) models
  • AIC and BIC
  • Predictions
  • AR series profile
• Moving Average (MA) models
  • MA series profile
• The ARMA model
• Seasonal series
  • Augmented Dicky-Fuller test
  • Transformations
  • Integration
• The ARIMA model
• Finding the best model
• Time series decomposition
• The SARIMA model
• Model diagnosis
• Calculation optimal orders
• The SARIMAX model
• Pandas library:
  • Temporal data types
  • Temporal data processing tools