The Electric Bill dataset contains monthly household electric billing charges for ten years. In addition, there are values for such potential explanatory variables as temperature, heating and cooling degree days, number in household, and indicator variables for a new electric meter and new heat pumps. The values provide a real dataset to use for applications ranging from simple graphical analysis through a variety of time series and causal forecasting methods. The dataset also is suited to spreadsheet applications for break-even calculations and optimization. With knowledge of the utility's tiered rate function, the bill amount can be converted to an estimate of the number of kilowatt hours used. A series of assignment questions is included and the accompanying Instructor's Manual provides solutions.
The CAUSE Research Group is supported in part by a member initiative grant from the American Statistical Association’s Section on Statistics and Data Science Education