Forecasting Seasonal Sales with Many Drivers: Shrinkage or Dimensionality Reduction?

↓ Direkt till sidans innehåll
↓ Direkt till sidans sekundära innehåll (sidomenyn)

Sökning: WFRF:(Fildes Robert) > Forecasting Seasona...

LIBRIS Formathandbok (Information om MARC21)

Fältnamn	Indikatorer	Metadata
000		04244naa a2200457 4500
001		oai:DiVA.org:his-22318
003		SwePub
008		230309s2023 \| \|\|\|\|\|\|\|\|\|\|\|000 \|\|eng\|
024	7	a https://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-223182 URI
024	7	a https://doi.org/10.3390/asi60100032 DOI
040		a (SwePub)his
041		a engb eng
042		9 SwePub
072	7	a ref2 swepub-contenttype
072	7	a art2 swepub-publicationtype
100	1	a Ramos, Patríciau Centre for Enterprise Systems Engineering, INESC TEC, Porto Accounting and Business School, Polytechnic of Porto, Asprela, Portugal4 aut
245	1 0	a Forecasting Seasonal Sales with Many Drivers :b Shrinkage or Dimensionality Reduction?
264		c 2022-12-26
264	1	b MDPI,c 2023
338		a electronic2 rdacarrier
500		a CC BY 4.0© 2022 by the authors.This research received no external funding.
520		a Retailers depend on accurate forecasts of product sales at the Store × SKU level to efficiently manage their inventory. Consequently, there has been increasing interest in identifying more advanced statistical techniques that lead to accuracy improvements. However, the inclusion of multiple drivers affecting demand into commonly used ARIMA and ETS models is not straightforward, particularly when many explanatory variables are available. Moreover, regularization regression models that shrink the model’s parameters allow for the inclusion of a lot of relevant information but do not intrinsically handle the dynamics of the demand. These problems have not been addressed by previous studies. Nevertheless, multiple simultaneous effects interacting are common in retailing. To be successful, any approach needs to be automatic, robust and efficiently scaleable. In this study, we design novel approaches to forecast retailer product sales taking into account the main drivers which affect SKU demand at store level. To address the variable selection challenge, the use of dimensionality reduction via principal components analysis (PCA) and shrinkage estimators was investigated. The empirical results, using a case study of supermarket sales in Portugal, show that both PCA and shrinkage are useful and result in gains in forecast accuracy in the order of 10% over benchmarks while offering insights on the impact of promotions. Focusing on the promotional periods, PCA-based models perform strongly, while shrinkage estimators over-shrink. For the non-promotional periods, shrinkage estimators significantly outperform the alternatives.
650	7	a SAMHÄLLSVETENSKAPx Ekonomi och näringslivx Företagsekonomi0 (SwePub)502022 hsv//swe
650	7	a SOCIAL SCIENCESx Economics and Businessx Business Administration0 (SwePub)502022 hsv//eng
653		a forecasting
653		a principal components analysis
653		a promotions
653		a retailing
653		a seasonality
653		a shrinkage
653		a Skövde Artificial Intelligence Lab (SAIL)
653		a Skövde Artificial Intelligence Lab (SAIL)
700	1	a Oliveira, José Manuelu Centre for Telecommunications and Multimedia, INESC TEC, Faculty of Economics, University of Porto, Porto, Portugal4 aut
700	1	a Kourentzes, Nikolaosu Högskolan i Skövde,Institutionen för informationsteknologi,Forskningsmiljön Informationsteknologi,Skövde Artificial Intelligence Lab (SAIL)4 aut0 (Swepub:his)koun
700	1	a Fildes, Robertu Centre for Marketing Analytics and Forecasting, Department of Management Science, Lancaster University Management School, United Kingdom4 aut
710	2	a Centre for Enterprise Systems Engineering, INESC TEC, Porto Accounting and Business School, Polytechnic of Porto, Asprela, Portugalb Centre for Telecommunications and Multimedia, INESC TEC, Faculty of Economics, University of Porto, Porto, Portugal4 org
773	0	t Applied System Innovationd : MDPIg 6:1q 6:1x 2571-5577
856	4	u https://doi.org/10.3390/asi6010003y Fulltext
856	4	u https://his.diva-portal.org/smash/get/diva2:1742322/FULLTEXT01.pdfx primaryx Raw objecty fulltext:print
856	4 8	u https://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-22318
856	4 8	u https://doi.org/10.3390/asi6010003

Hitta via bibliotek

Applied System Innovation (Sök värdpublikationen i LIBRIS)

Till lärosätets databas

Hitta mer i SwePub

Av författaren/redakt...: Ramos, Patrícia; Oliveira, José M ...; Kourentzes, Niko ...; Fildes, Robert

Om ämnet

SAMHÄLLSVETENSKAP: SAMHÄLLSVETENSKA ...; och Ekonomi och näri ...; och Företagsekonomi

Artiklar i publikationen: Applied System I ...

Av lärosätet: Högskolan i Skövde

Sök utanför SwePub

Sök vidare i:: Google; Google Book Search; Google Scholar

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.

LIBRIS.kb.se