Choosing a method for predicting economic performance of companies

Show simple item record
dc.contributor.authorDvořáček, Jaroslav
dc.contributor.authorSousedíková, Radmila
dc.contributor.authorŘepka, Michal
dc.contributor.authorDomaracká, Lucia
dc.contributor.authorBarták, Pavel
dc.contributor.authorBartošíková, Michaela
dc.date.accessioned2012-07-31T07:52:45Z
dc.date.available2012-07-31T07:52:45Z
dc.date.issued2012
dc.description.abstractThis paper reports on the eff orts to fi nd a method for predicting economic results of companies. The input data fi les consist of 93 profi table companies and 93 bankrupt fi rms. From the total number of 93 fi rms in both categories, data of 72 fi rms served for establishing a classifi cation criterion, and for the rest of 21 fi rms, a prognosis of their economic development was performed. The classifi cation criterion for prognosticating the future economic development has been established by applications of discriminate analysis, logit analysis, and artifi cial neural network methods. The application of artifi cial neural networks has provided for better classifi cation accuracies of 90,48 % for successful fi rms, and 100 % for bankrupt fi rms.cs
dc.description.abstract-enU ovom se članku opisuju pokušaji pronalaska učinkovite metode za predviđanje ekonomskog rezultata tvrtki. Datoteke ulaznih podataka sastoje se od 93 uspješne tvrtke i 93 tvrtke koje su bankrotirale. Od ukupnog broja od 93 tvrtke u obje kategorije datoteka s ulaznim podacima, podaci za 72 tvrtke poslužili su za određivanje klasifi kacijskog kriterija a za preostalu 21 tvrtku provela se prognoza njihovog ekonomskog razvoja. Klasifi kacijski kriterij za predviđanje budućeg ekonomskog razvoja tvrtke uspostavljen je primjenom analize diskriminacije, logičkom regresijom i metodama umjetne neuronske mreže. Primjena logičke regresije i umjetnih neuronskih mreža omogućila je bolju klasifi kacijsku točnost u slučaju 90,48 % uspješnih tvrtki i 100 % tvrtki koje su bankrotirale.cs
dc.description.firstpage525cs
dc.description.issue4cs
dc.description.lastpage528cs
dc.description.sourceWeb of Sciencecs
dc.description.volume51cs
dc.format.extent958112 bytescs
dc.format.mimetypeapplication/pdfcs
dc.identifier.citationMetalurgija. 2012, g. 51, br. 4, s. 525-528.cs
dc.identifier.issn0543-5846
dc.identifier.issn1334-2576
dc.identifier.locationNení ve fondu ÚKcs
dc.identifier.urihttp://hdl.handle.net/10084/94932
dc.identifier.uri
dc.identifier.wos000305205800023
dc.language.isoencs
dc.publisherHrvatsko Metalurško Društvocs
dc.relation.ispartofseriesMetalurgijacs
dc.relation.urihttp://hrcak.srce.hr/file/123401cs
dc.rights.accessopenAccess
dc.subjectpredictioncs
dc.subjectdiscriminant analysiscs
dc.subjectLogit analysiscs
dc.subjectartificial neural networkscs
dc.subjectclassification accuraciescs
dc.subjectpredvidljivostcs
dc.subjectanaliza razlikacs
dc.subjectanalize logičke regresije (Logit analize)cs
dc.subjectumjetne neuronske mrežecs
dc.subjectklasifkacija točnostics
dc.titleChoosing a method for predicting economic performance of companiescs
dc.title.alternativeIzbor metode za predviđanje ekonomskih rezultata tvrtkics
dc.typearticlecs
dc.type.statusPeer-reviewedcs
dc.type.versionpublishedVersioncs

Files

Original bundle

Now showing 1 - 1 out of 1 results
Thumbnail Image
Name:
MET_51_4_525_528_Dvoracek.pdf
Size:
935.66 KB
Format:
Adobe Portable Document Format
Description:
publishedVersion
Download

License bundle

Now showing 1 - 1 out of 1 results
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description:
Download

Collections

Publikační činnost VŠB-TUO ve Web of Science / Publications of VŠB-TUO in Web of Science
OpenAIRE
Publikační činnost Katedry ekonomiky a systémů řízení / Publications of Department of Economics and Control Systems (545)
Články z časopisů s impakt faktorem / Articles from Impact Factor Journals