Decomposing time series data via mixed integer programming
dc.contributor.author | Gözüyılmaz, Şeyma | |
dc.date.accessioned | 2020-03-20T09:08:40Z | |
dc.date.available | 2020-03-20T09:08:40Z | |
dc.date.issued | 2020-01-13 | |
dc.identifier.uri | http://hdl.handle.net/10679/6422 | |
dc.identifier.uri | https://tez.yok.gov.tr | |
dc.identifier.uri | http://discover.ozyegin.edu.tr/iii/encore/record/C__Rb3984697?lang=eng | |
dc.description | Thesis (M.A.)--Özyeğin University, Graduate School of Sciences and Engineering, Department of Industrial Engineering, February 2020. | |
dc.description.abstract | Decomposing time series into seasonality, trend, and remainder reveals underlying insights to be used in forecasting and anomaly detection. Although there are several decomposition methods, no method guarantees all of the following issues are addressed: i) smoothness of trend and the rigid structure of seasonality, ii) shifts in trend, iii) long seasonality periods, iv) multi-seasonality, and v) robustness on outliers. In this study, we propose a mixed integer programming model to address all of these issues. Experiments on di↵erent synthetic problem sets present the e↵ectiveness of the proposed algorithm, providing benchmark results against the robust seasonal trend decomposition algorithm. | en_US |
dc.description.abstract | Zaman serilerini trend, sezonsallık ve arta kalan olarak ayırmak, tahmin yapmada ve anormallik belirlemede kullanılacak temelindeki i¸cg¨or¨uleri ortaya ¸cıkarmaktadır. Bir¸cok ayrı¸stırma y¨ontemi olmasına ra˘gmen, hi¸cbir y¨ontem takip eden konuların hepsini ele alaca˘gını garanti etmemektedir. Bu konular i) trendin d¨uzg¨unl¨u˘g¨u ve sezonsallı ˘gın katı yapısı ii) trend’deki de˘gi¸simler iii) uzun sezonsallık d¨onemleri iv) ¸coklu sezonsallık ve v) u¸c de˘gerlerdeki g¨urb¨uzl¨ukt¨ur. Bu ¸calı¸smada, t¨um bu konuları ele alabilmek adına bir tam sayı programlama modeli ¨oneriyoruz. Farklı sentetik problem k¨umeleri ¨uzerinde yapılan deneyler, ¨onerilen algoritmanın etkilili˘gini ve g¨urb¨uz sezonsallık trend ayrı¸stırma algoritmasına kar¸sılık de˘gerlendirme sonu¸clarını ortaya koymaktadır. | |
dc.language.iso | eng | en_US |
dc.rights | restrictedAccess | |
dc.title | Decomposing time series data via mixed integer programming | en_US |
dc.title.alternative | Zaman serilerinin karmaşık tam sayılı programlama ile parçalarına ayrıştırılması | |
dc.type | Master's thesis | en_US |
dc.contributor.advisor | Kundakçıoğlu, Ömer Erhun | |
dc.contributor.committeeMember | Kundakçıoğlu, Ömer Erhun | |
dc.contributor.committeeMember | Albey, Erinç | |
dc.contributor.committeeMember | Baydoğan, M. G. | |
dc.publicationstatus | Unpublished | en_US |
dc.contributor.department | Özyeğin University | |
dc.contributor.ozugradstudent | Gözüyılmaz, Şeyma | |
dc.contributor.authorFemale | 1 | |
dc.relation.publicationcategory | Thesis - Institutional Graduate Student |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Master's Theses
This Collection covers master's thesis produced at Özyeğin University
Share this page