![\"\"](\"http://www.caos.inf.uc3m.es/%7Ejperalta/ICTSF/images/Evaluation.PNG\")
Remember, the competition deadline is the 10th of February. Participants must write a IEEE pdf 6 pages paper explaining the system designed, attach the excel with the forecast of each time series and send it to\r\njperalta@inf.uc3m.es.
\r\nOn the other hand, EAIS special session submission is the 15th of February. Participants can go to easychair submission (https://www.easychair.org/account/signin.cgi?conf=cifts2012)\r\n and submit the same pdf document used for the competition as a contribution paper.
\r\nFor more information go to http://www.caos.inf.uc3m.es/~jperalta/ICTSF/ or email\r\njperalta@inf.uc3m.es.
\r\n\r\n
The International Competition of Time Series Forecasting is a forecasting competition which aims to encourage the development of new models to evaluate the accuracy of computational intelligence (CI) methods in time series forecasting with multiple frequencies. The\r\n contestants will use a unique consistent methodology for all time series.
\r\nForecast a dataset of 8 time series from a web bookshop data as accurately as possible, using methods from computational intelligence and applying a consistent methodology. The data consists of 8 time series with\r\n different time frequencies, that could include yearly, quarterly, monthly, weekly, daily and hourly data.
\r\nThe prediction competition is open to all methods of computational intelligence, incl. feed-forward and recurrent neural networks, fuzzy predictors, evolutionary & genetic algorithms, decision & regression tress, support vector\r\n regression, hybrid approaches etc. used in all areas of forecasting, prediction & time series analysis, etc.
\r\n\r\nYou will be able to find in data an example_submission.csv file. It currently contains the last observed value for each time series as the prediction for each of the remaining values, update this value with your own predictions\r\n and submit it.
\r\n
\r\n
\r\n\r\nTo participate, first send an email to \r\njperalta@inf.uc3m.es to confirm you are interested in participating in this competition.
\r\n\r\n
\r\n\r\n
\r\n\r\nAll submissions will be invited to participate into\r\n\r\nIEEE-EAIS 2012 conference, in the special session titled "Computational Intelligence applied to Forecast Time Series".\r\n For more information contact \r\njperalta@inf.uc3m.es
\r\n\r\n
\r\n\r\nFor further information go to: \r\nhttp://www.caos.inf.uc3m.es/~jperalta/ICTSF/
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Evaluation", "content": "\r\nThe global performance of a forecasting model is evaluated by an error measure. Historically, Mean Absolute Error (MAE) or (Root) Mean Squared Error ((R)MSE) are very popular error measures. However, Mean Square Error is too sensitive to outliers [1] and furthermore,\r\n both (R)MSE and MAE are scale-dependent measures and hence, it can be hardly used for a comparison across more time series since every single time series has a different impact on the overall results [2]. For example, it has been shown that five of the 1001\r\n series from the M-competition dominated the RMSE ranking of the forecasting methods and the remaining 996 series had only little impact on the ranking [1].
\r\n\r\nSymmetric Mean Absolute Percentage Error (SMAPE) [3]:
\r\n\r\n
\r\n\r\n
\r\n
\r\n\r\nwhere et = yt-Ft for t = T +1, ..., T +h, it belongs to scale independent error measures, thus can be more easily used to compare methods across different time series. Although the SMAPE was originally proposed in [4] in a different\r\n form, formula (10) adopts the variant used in [5] since it does not lead to negative values (ranging from 0% to 200%).
\r\n\r\nFor the sake of correct evaluation, we choose SMAPE to measure the errors of the competition.
\r\n\r\n
\r\n\r\nSelected references
\r\n\r\nJ.S. Armstrong, F. Collopy, Error measures for generalizing about forecasting methods: Empirical comparisons, International Journal of Forecasting 8 (1992) 69-80.
\r\n\r\nJ.S. Armstrong, Evaluating methods, in: J.S. Armstrong (Eds.), Principles of Forecasting: A handbook for reasearchers and practitioners, Chap. 14, Kluwer, Boston/Dordrecht/London, 2001, pp. 443-473.
\r\n\r\nR. J. Hyndman, A. Koehler, Another look at measures of forecast accuracy, International Journal of Forecasting 22(4) (2006) 679-688.
\r\n\r\nJ.S. Armstrong, Long-range forecasting, Wiley New York ETC., 1985.
\r\n\r\nR. Andrawis, A. Atiya, A new Bayesian formulation for Holt's exponential smoothing, Journal of Forecasting 28(3) (2009) 218-234.
\r\nIn Data folder participants will be able to see different files:
\r\nTo carry out a submission, participants just have to open sample_entry.csv file, change the last column for the values they obtain as prediction for each time series, go to Submission folder and see how good your system is.
\r\nCareful!!, you will be able to check the accuracy of your system only once per day. GOOD LUCK!!
", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false }, { "name": "Rules", "content": "\r\n\t\r\nYou cannot sign up to Kaggle from multiple accounts and therefore you cannot submit from multiple accounts.
\r\n\r\n Privately sharing code or data outside of teams is not permitted.\r\n It's okay to share code if made available to all participants on the forums.\r\n
\r\n\r\n Team mergers are allowed and can be performed by the team leader. In order to merge, the combined team must have a total submission count less than or equal to the maximum allowed as of the merge date. The maximum allowed is the number of submissions per day multiplied by the number of days the competition has been running.
\r\n
There is no maximum team size.
\r\nYou may submit a maximum of 1 entries per day.
\r\nYou may select up to 5 final submissions for judging.
\r\nStart Date: 11/21/2011 1:05 PM UTC
\r\nMerger Deadline: None
\r\nEntry Deadline: None
\r\nEnd Date: 2/10/2012 11:59 PM UTC
\r\n\t\r\n", "isPublished": false, "hasIsPublished": false, "postTitle": "", "hasPostTitle": false, "mimeType": "", "hasMimeType": false } ] }