Outliers in regression modelling: Influential vs. non-influential values and detection using information criteria
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27510%2F19%3A10243466" target="_blank" >RIV/61989100:27510/19:10243466 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Outliers in regression modelling: Influential vs. non-influential values and detection using information criteria
Original language description
In the estimation of regression models in urban valuation, the detection of atypical values is of great importance to avoid possible spurious results, as a small subset of these observations, can exert a high influence in the parameters estimates. The Akaike Information Criterion is used to characterize multivariate outliers in non-robust regression modelling. The discriminating power to detect those outliers that are influential observations is analyzed, obtaining better results that with the classic methods available in well-known statistical packages in regression. This is of great importance in the construction of estimation models. A simulation modelling is performed to assess the validity of the proposed procedure in contrast with classical outlier detection methods in regression. The use of a Monte Carlo simulation study is motivated on the difficulties in the analysis of the sampling distribution of the AIC.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
50202 - Applied Economics, Econometrics
Result continuities
Project
<a href="/en/project/EE2.3.20.0296" target="_blank" >EE2.3.20.0296: Research team for modelling of economic and financial processes at VSB-TU Ostrava</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach
Others
Publication year
2019
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Proceedings of the 13th International Conference on Strategic Management and its Support by Information Systems: May 21th-22th, 2019, Ostrava, Czech Republic
ISBN
978-80-248-4305-6
ISSN
2570-5776
e-ISSN
—
Number of pages
12
Pages from-to
261-272
Publisher name
VŠB - Technical University of Ostrava
Place of publication
Ostrava
Event location
Ostrava
Event date
May 21, 2019
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—