City of London Water: Predicting Electricity Prices and Optimizing Operations
(4 pages of text)
Case (Pub Mat)
The manager of Water Operations at the City of London wanted to reduce the cost of pumping water by optimizing energy use. Water Operations had been consistently experiencing higher operational costs, and electricity bills accounted for up to 30 per cent of its direct costs. If the manager could predict the future cost of energy, she could schedule the operation of the water pumps to benefit from periods of low energy prices, potentially saving up to 25 per cent of the energy costs. The manager obtained the hourly Ontario electricity prices from 2003 to 2015, and wanted to use that data to evaluate how accurately she could predict future electricity prices. How should she proceed?
The case is suitable for use in an undergraduate- or graduate-level course on data analytics or finance. Previous knowledge of regression models, autoregressive models, and moving average models is required for the case. It introduces and discusses the value of forecasting and time series analysis; and seasonality, stationarity, and seasonal autoregressive integrated moving average (SARIMA) models. After working through the case and assignment questions, students will be able to
- determine stationarity;
- determine parameters for SARIMA models;
- create forecasts; and
- determine the validity of the models.
$4.25 CAD / $4.25 USD Printed Copy
$3.75 CAD / $3.75 USD Permissions
$3.75 CAD / $3.75 USD Digital Download