The assumptions log documents inputs and assumptions, giving information about the source, methodology, update dates and sign-off process. It also allows teams to identify assumptions which may carry higher risk. A good assumptions log will help models score highly on the criteria linked to Data & Assumptions in DECC’s Quality Assurance log.
This assumptions log template is used by model-owners within DECC, and it should also be used for all new models which are procured by DECC.
Logging assumptions correctly is an important part of the Quality Assurance of any model. It ensures that there is a clear justification for decisions made about which data and methodologies are used in model.
An assumption might be a number (e.g. a historical data point or series, a projected data point or series, a conversion or efficiency factor, a technology parameter) or a methodological assumption (e.g. a method for extrapolation, a behavioural assumption, a statistical assumption, a choice of calculation methodology).