Choosing the number of conditioning events in judgemental forecasting
Article Abstract:
The conditioning events play a major role in the accuracy of a judgemental forecast. An analysis shows that conditioning events with precise estimation yield proves to be beneficial in forecasting than conditioning events that are informative. On the other hand, the need to increase the number of conditioning events in a judgemental forecasting is highly dependent on the estimation accuracy of the marginal probabilities of conditioning events. Finding also shows that a slightly larger number of conditioning events is desirable in forecasting than using too many conditioning events.
Publication Name: Journal of Forecasting
Subject: Mathematics
ISSN: 0277-6693
Year: 1997
User Contributions:
Comment about this article or add new information about this topic:
Going up-going down: how good are people at forecasting trends and changes in trends?
Article Abstract:
A laboratory-based study designed to determine whether the direction of time series affects forecast accuracy was conducted. The study, which presented controlled experiment of series direction, also investigated the problems of changing trends. Results revealed that problems with downward-sloping series are usually encountered, since they are forecast badly and with a possibility towards dampening. Forecast accuracy was measured in terms of 'smoothed' absolute percentage error.
Publication Name: Journal of Forecasting
Subject: Mathematics
ISSN: 0277-6693
Year: 1997
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Asymmetry in the link between the yield spread and industrial production: threshold effects and forecasting. Density forecasting in economics and finance
- Abstracts: On the revelation of private information in stock market economies. The graphs of the Walras correspondence: the production economies case
- Abstracts: Autoregressive gamma processes. Stochastic models underlying CrostonEs method for intermittent demand forecasting
- Abstracts: Prediction intervals for growth curve forecasts. Order series method for forecasting non-Gaussian time series
- Abstracts: Time series multistep-ahead predictability estimation and ranking. Predictions in overdispersed series of counts using an approximate predictive likelihood