A simulation study for count data models under varying degrees of outliers and zeros


Tuzen F., Erbas S., OLMUŞ H.

COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION, cilt.49, ss.1078-1088, 2020 (SCI İndekslerine Giren Dergi) identifier identifier

  • Cilt numarası: 49 Konu: 4
  • Basım Tarihi: 2020
  • Doi Numarası: 10.1080/03610918.2018.1498886
  • Dergi Adı: COMMUNICATIONS IN STATISTICS-SIMULATION AND COMPUTATION
  • Sayfa Sayıları: ss.1078-1088

Özet

This study was aimed at examining the performance of count data models under various outliers and zero inflation situations with simulated data. Poisson, Negative Binomial, Zero-inflated Poisson, Zero-inflated Negative Binomial, Poisson Hurdle and Negative Binomial Hurdle models were considered to test how well each of the model fits the selected datasets having outliers and excess zeros. We found that Zero-inflated Negative Binomial and Negative Binomial Hurdle models were found to be more successful than other count data models. Also the results indicated that in some scenarios, the Negative Binomial model outperformed other models in the presence of outliers and/or excess zeros.