Making progress in forecasting

Abstract

Twenty-five years ago, the International Institute of Forecasters was established "to bridge the gap between theory and practice. " Its primary vehicle was the Journal of Forecasting and is now the International Journal of Forecasting. The Institute emphasizes empirical comparisons of reasonable forecasting approaches. Such studies can be used to identify the best forecasting procedures to use under given conditions, a process we call evidence-based forecasting. Unfortunately, evidence-based forecasting meets resistance from academics and practitioners when the findings differ from currently accepted beliefs. As a consequence, although much progress has been made in developing improved forecasting methods, the diffusion of useful forecasting methods has been disappointing. To bridge the gap between theory and practice, we recommend a stronger emphasis on the method of multiple hypotheses and on invited replications of important research. It is then necessary to translate the findings into principles that are easy to understand and apply. The Internet and software provide important opportunities for making the latest findings available to researchers and practitioners. Because researchers and practitioners believe that their areas are unique, we should organize findings so that they are relevant to each area and make them easily available when people search for information about forecasting in their area. Organisational barriers to change still remain to be overcome. Research into the specific issues faced when forecasting remains a priority. KeywordsBarriers to implementation, Evidence-based forecasting, forecasting practice, forecasting software, freeware, replication should organize findings so that they are relevant to each area and make them easily available when people search for information about forecasting in their area. Organisational barriers to change still remain to be overcome. Research into the specific issues faced when forecasting remains a priority. Comments Forthcoming in the KeywordsBarriers to implementation; Evidence-based forecasting; forecasting practice; forecasting software; freeware; replication.In 1980 . These were exciting days. We viewed ourselves as visionaries and set an ambitious objective: "To unify the field and to bridge the gap between theory and practice". We planned to publish research to identify the best procedures for forecasting. In this paper, we look back at our successes and failures in the last 25 years and suggest ways to accelerate future progress. Foundations of ProgressIn 1890, Chamberlin, a geologist and later president of the American Association for the Advancement of Science, examined why some areas of science progress rapidly while others do not. He concluded that those sciences that test multiple reasonable hypotheses were the ones that made the most progress.A properly conducted test of multiple reasonable hypotheses about an important problem should always be of some value: it can tell which hypotheses (or methods) are most appropriate under given...

Forecasting Daily Patient Volumes in the Emergency Department

“…Many methods have been developed to model and forecast time series, making the choice of an appropriate method an important task. Extensive research conducted by forecasters indicates that there is no universally superior forecasting method and that theoretical arguments can be made in favor of multiple methods given the same set of data 4,22,23 . Therefore, good forecasting practice suggests that multiple forecasting methods should be compared and evaluated based on their ability to forecast postsample observations 23 .…”

Validation and forecasting accuracy in models of climate change

“…One fundamental contribution of forecasting research is its emphasis on the requirement that a method (or forecasting process) demonstrates its superiority by beating some plausible competing benchmark. In so far as researchers know how to select a good forecasting method ex ante, perhaps the primary requirement is that it must have been shown to work previously in circumstances similar to those which are expected to apply in the future, outperforming the alternatives, and in particular a benchmark (Armstrong & Fildes, 2006). Of course, it is expected that in small samples, the noise may well overwhelm the signal (in the GCMs derived from increasing CO 2 emissions and concentration levels), and therefore a large sample of forecasts may need to be considered.…”

Using a Software Tool in Forecasting: a Case Study of Sales Forecasting Taking into Account Data Uncertainty

“…Wang et al [4] proposed residual modification models to improve the precision of seasonal ARIMA for electricity demand forecasting. The forecasting based on hybrid methodology that combines both autoregressive integrated moving average (ARIMA) and artificial neural network (ANN) models for predicting was presented in [5][6][7][8]. Kavasseri and Seetharaman [9] examined the use of fractional-ARIMA or f-ARIMA models to model, and forecast wind speeds.…”