Retropolation
Retropolation, or retrospective forecasting, is the reverse of conventional forecasting: it estimates unknown values at the beginning of a time series using known values from later periods.
Retropolation can be used to study past conditions based on present-day evidence or to plan current actions by reasoning backward from a desired future state.
As a research method, retropolation uses current and recent trends in a system to infer its earlier behavior. As a planning method, this reverse-time approach—often called backcasting—starts with a desired future outcome and works backward to identify the policies and programs needed to connect that outcome to the present.
The backcasting approach described by John B. Robinson in 1982 begins with a specific, clearly defined future scenario. The analyst then works backward through as many steps as necessary to reach the present and identify a plausible sequence of actions that could make that future scenario achievable.