Combining forecasts has been shown in many cases to lead to improvements in forecasting performance, in terms of accuracy and bias. This is also common in forecasting with neural networks or other computationally intensive methods, where ensemble forecasts are considered more accurate than individual model forecasts. A useful feature of forecast combination is that it… Read More »
Although Croston’s method and its variants are popular for intermittent demand time series, there have been limited advances in identifying how to select appropriate smoothing parameters and initial values. From the one hand this complicates forecasting for organisations, and from the other hand it does not permit automation. Recent research investigated various cost functions for… Read More »
I uploaded today to CRAN an updated version of tsintermittent. Yesterday I found a bug in the optimisation of the Croston’s, SBA and TSB methods and this update fixes that. Here is the changelog for this update: Version 1.3 (04 August 2014) * Corrected major optimisation bug – issue with opt.on in all crost, tsb… Read More »
Shiny is a web application framework for R. The idea is simple: deploy R code in webpages. This might prove useful when user interaction is required, for instance to design and deploy forecasting experiments that need human participants. I gave it a try to see how easy is it to build a demo. Assuming your… Read More »
Want to learn more about the Forecasting Society? These videos will give you an idea what it is all about. Are you a researcher in forecasting? or are you interested in participating in forecasting research? Either way, if you are interested in the area and in particular when human judgement is involved, the Forecasting Society… Read More »
Recently I posted about a paper I co-authored with Fotios Petropoulos, now in JORS: Forecast Combinations for Intermittent Demand. There we found that for intermittent demand data using multiple levels of temporal aggregation, forecasting them with the appropriate models and finally combining the forecasts performed best. This approach has many analogies with the MAPA algorithm… Read More »
Over the years I have reviewed numerous papers that do not properly benchmark the various methods proposed. In my opinion if a paper has an empirical evaluation, then it must have appropriate benchmarks as well. Otherwise, one cannot claim that convincing empirical evidence is provided. The argument is simple: if the proposed method does not… Read More »
Together with Fotios Petropoulos we launched a new portal to facilitate judgemental forecasting research. The aim of the Forecasting Society (www.forsoc.net) is to bring together researchers in judgemental forecasting and participants from practice and academia. At the same we hope that it will grow to be a discussion forum to exchange and promote judgemental forecasting… Read More »
A new package for analysing and forecasting intermittent demand time series and slow moving items has been release for R. You can download the latest version from CRAN. The launch version contains the following functions: crost: Croston’s method and variants. crost.ma: Moving average with Croston’s method decomposition. idclass: Time series categorisation for intermittent demand. simID:… Read More »
N. Kourentzes, 2014, International Journal of Production Economics, 156: 180-190. http://dx.doi.org/10.1016/j.ijpe.2014.06.007