Multivariate Time Series Forecasting: Methods and Applications for Wastewater Quality Prediction.

Abstract:

I will begin by recalling the main concepts of multivariate time series analysis as well as modeling approaches for forecasting, in particular the Seasonal Autoregressive Integrated Moving Average (SARIMA) model in a vector framework and Long Short-Term Memory (LSTM) neural networks. I will then present their application to the forecasting of several variables characterizing wastewater quality (pH, temperature, flow rate, etc.).