Datum
2020Metadata
Zur Langanzeige
Aufsatz
The Impact of Missing Values on PLS, ML and FIML Model Fit
Zusammenfassung
Structural equation modelling has become widespread in the marketing research domain due to the possibility of creating and investigating latent constructs. Today, several estimation methods are available, each with strengths and drawbacks. This study investigates how the established estimation methods of partial-least-squares (PLS), maximum likelihood (ML) and full-information maximum likelihood (FIML) perform with an increasing percentage of missing values (MVs). The research was conducted by investigating an adapted model of the European customer satisfaction index (ECSI). MVs were randomly generated with an algorithm. The performance of PLS, ML and FIML was tested with eight data sets that contained between 2.22% and 27.78% randomly generated MVs. It was shown that ML performs relatively poorly if the percentage of MVs exceeds 7%, while PLS performs satisfactorily if the percentage of MVs does not exceed 9%. FIML was shown to be mostly stable up to 17% MVs.
Zitierform
In: Archives of Data Science, Series A Vol. 6 / No. 1 (2020) issn:2363-9881Zitieren
@article{doi:10.17170/kobra-202107074257,
author={Grimm, Malek Simon and Wagner, Ralf},
title={The Impact of Missing Values on PLS, ML and FIML Model Fit},
journal={Archives of Data Science, Series A},
year={2020}
}
0500 Oax 0501 Text $btxt$2rdacontent 0502 Computermedien $bc$2rdacarrier 1100 2020$n2020 1500 1/eng 2050 ##0##http://hdl.handle.net/123456789/13142 3000 Grimm, Malek Simon 3010 Wagner, Ralf 4000 The Impact of Missing Values on PLS, ML and FIML Model Fit / Grimm, Malek Simon 4030 4060 Online-Ressource 4085 ##0##=u http://nbn-resolving.de/http://hdl.handle.net/123456789/13142=x R 4204 \$dAufsatz 4170 5550 {{Maximum-Likelihood-Schätzung}} 5550 {{Verbraucherzufriedenheit}} 5550 {{Marktforschung}} 5550 {{Methode}} 7136 ##0##http://hdl.handle.net/123456789/13142
2021-08-20T13:37:47Z 2021-08-20T13:37:47Z 2020 doi:10.17170/kobra-202107074257 http://hdl.handle.net/123456789/13142 eng Namensnennung - Weitergabe unter gleichen Bedingungen 4.0 International http://creativecommons.org/licenses/by-sa/4.0/ 330 The Impact of Missing Values on PLS, ML and FIML Model Fit Aufsatz Structural equation modelling has become widespread in the marketing research domain due to the possibility of creating and investigating latent constructs. Today, several estimation methods are available, each with strengths and drawbacks. This study investigates how the established estimation methods of partial-least-squares (PLS), maximum likelihood (ML) and full-information maximum likelihood (FIML) perform with an increasing percentage of missing values (MVs). The research was conducted by investigating an adapted model of the European customer satisfaction index (ECSI). MVs were randomly generated with an algorithm. The performance of PLS, ML and FIML was tested with eight data sets that contained between 2.22% and 27.78% randomly generated MVs. It was shown that ML performs relatively poorly if the percentage of MVs exceeds 7%, while PLS performs satisfactorily if the percentage of MVs does not exceed 9%. FIML was shown to be mostly stable up to 17% MVs. open access Grimm, Malek Simon Wagner, Ralf doi:10.5445/KSP/1000098011/04 Maximum-Likelihood-Schätzung Verbraucherzufriedenheit Marktforschung Methode publishedVersion issn:2363-9881 No. 1 Archives of Data Science, Series A Vol. 6 false
Die folgenden Lizenzbestimmungen sind mit dieser Ressource verbunden: