Data: Boundaries and Assumptions

Joaquin Narro and Monica Caamano

We have explored several aspects of systematising models, including practical examples, focusing on broad defining lines that will help the reader develop their own systematic models. We are now ready to dig further into the detail by investigating data-related model boundaries and assumptions.

In general, data can be sorted into technical (price) and fundamental data. Although the range of data-related issues is very extensive, we will focus on practical aspects associated with data quality and availability, exploring the practicalities of using weather data and settlement data as two specific energy-related concerns, starting with the establishment of a data quality framework.

THE NEED FOR A DATA QUALITY FRAMEWORK

In any circumstances, whether dealing with price or fundamental data, we would suggest using a framework for evaluating data quality that monitors completeness, credibility, precision and continuity (Cichy and Rass, 2019).

• • Completeness: This describes the percentage of data available. For example, if we are looking at daily settlement prices, we need to know how many days of data are missing from the time series.

• • Credibility: This