Using inferred probabilities to measure the accuracy of imprecise forecasts

Using inferred probabilities to measure the accuracy of imprecise forecasts

Show other versions (1)

Research on forecasting is effectively limited to forecasts that are expressed with clarity; which is to say that the forecasted event must be sufficiently well-defined so that it can be clearly resolved whether or not the event occurred and forecasts certainties are expressed as quantitative probab...

Full description

Bibliographic Details
Main Authors: Paul Lehner, Avra Michelson, Leonard Adelman, Anna Goodman
Format: Article
Language:English
Published: Cambridge University Press 2012-11-01
Series:Judgment and Decision Making
Subjects:
Research on forecasting is effectively limited to forecasts that areexpressed with clarity; which is to say that the forecasted event mustbe sufficiently well-defined so that it can be clearly resolvedwhether or not the event occurred and forecasts certainties areexpressed as quantitative probabilities. When forecasts are expressedwith clarity
then quantitative measures (scoring rules
calibration
discrimination
etc.) can be used to measure forecast accuracy
whichin turn can be used to measure the comparative accuracy of differentforecasting methods. Unfortunately most real world forecasts are notexpressed clearly. This lack of clarity extends to both thedescription of the forecast event and to the use of vague language toexpress forecast certainty. It is thus difficult to assess theaccuracy of most real world forecasts
and consequently the accuracythe methods used to generate real world forecasts. This paperaddresses this deficiency by presenting an approach to measuring theaccuracy of imprecise real world forecasts using the same quantitativemetrics routinely used to measure the accuracy of well-definedforecasts. To demonstrate applicability
the InferredProbability Method is applied to measure the accuracy of forecasts infourteen documents examining complex political domains.Key wordsforecasting
forecast accuracy
imprecise forecasts
politicalforecasting
verbal probability
probability calibration.
Online Access:http://journal.sjdm.org/10/101116/jdm101116.pdf

Internet

http://journal.sjdm.org/10/101116/jdm101116.pdf

Similar Items