Araştırma Makalesi
BibTex RIS Kaynak Göster

A Comparison on Count Data Models: Factors Affecting the Number of Houses Example of Türkiye

Yıl 2024, Cilt: 17 Sayı: 1, 30 - 35, 20.11.2024
https://doi.org/10.58688/kujs.1467396

Öz

Poverty is a multidimensional concept. One of the indicators of poverty is the number of houses owned by the household. In this study, counting regression models were used to determine the factors affecting the number of houses owned by household members. Moreover, it was investigated which regression model would best fit the data. The most commonly used count regression models are classical count regression models (such as Poisson and negative binomial) and zero truncated regression models. However, another count regression model proposed in the literature is zero-truncated count regression models. These models prevent the loss of time and cost caused by analyzing all the data when there is a desired range in the data. Therefore, these models are a good option to use in modeling situations where count data is available. Various count regression models were applied to the Income and Living Conditions Survey data set by TURKSTAT. The performance evaluation of the models considered in the study was made. Akaike Information Criterion and Log Likelihood value were used to compare the suitability of the models. As a result, the zero-truncated negative binomial regression model is the model that best fits the real data set.

Kaynakça

  • Agresti A. (2002). Categorical data analysis (Second Edition), New Jersey: Wiley & Sons Incorporation.
  • Akaike, H. (1973). Information theory and extension of the maximum likelihood principle, Second International Symposium on Information Theory, Budapest: Akademiai Kiado, 267–281.
  • Altun, E. (2018). A new zero-inflated regression model with application. İstatistikçiler Dergisi: İstatistik ve Aktüerya, 11(2), 73-80.
  • Alwani, Z. Z., Ibrahim, A. I. N., Yunus, R. M., & Yusof, F. (2021). Application of zero-truncated count data regression models to air-pollution disease. In Journal of Physics: Conference Series (Vol. 1988, No. 1, p. 012096). IOP Publishing.
  • Cameron, A. C., and Trivedi, P. K. (2013). Regression analysis of count data (Second edition). New York: Cambridge university press, 128-132.
  • Durmuş, B., & Güneri, Ö. İ. (2020). An application of the generalized poisson model for over dispersion data on the number of strikes between 1984 and 2017. Alphanumeric Journal, 8(2), 249-260.
  • Gevrekci, Y., Guneri, O. I., Takma, C., & Yesilova, A. (2022). Comparison of different count models for investigation of some environmental factors affecting stillbirth in holsteins. Indian Journal of Animal Research, 56(9), 1158-1163.
  • Greene, W. H., (1994). Accounting for excess zeros and sample selection in poisson and negative binomial regression models. New York University Working Paper, 94(10), 1- 37.
  • Grogger, J. T., and Carson, R. T. (1991). Models for truncated counts. Journal of applied econometrics, 6(3), 225-238.
  • Hilbe, J. M. (2014). Modelling Count Data. New York: Cambridge University Press, 20- 170.
  • İnternet: https://data.tuik.gov.tr/Bulten/Index?p=Gelir-ve-Yasam-Kosullari-Arastirmasi2018-30755. Son erişim tarihi: 25.09.2020.
  • İşçi Güneri, Ö., & Durmuş, B. (2021). Models for Overdispersion Count Data with Generalized Distribution: An Application to Parasites Intensity. Journal of New Theory(35), 48-61. https://doi.org/10.53570/jnt.902066
  • Kim, D. W., Deo, R. C., Park, S. J., Lee, J. S., and Lee, W. S. (2019). Weekly heat wave death prediction model using zero-inflated regression approach. Theoretical and Applied Climatology, 137(1-2), 823-838.
  • Lawal, B. H. (2022). Zero-Truncated Models applied to the Nigerian National Health Insurance Data. BENIN JOURNAL OF STATISTICS , Vol. 5, pp. 1– 20.
  • Min, Y., and Agresti, A. (2005). Random effect models for repeated measures of zeroinflated count data. Statistical modelling, 5(1), 1-19.
  • Tüzen, M., F. and Erbaş, S. (2017). A comparison of count data models with an application to daily cigarette consumption of young persons. Communıcatıons In Statistics Theory And Methods, 47(23), 5825-5844.
  • Worku, G., Tadesse, G., Arega, A., & Tesfaw, D. (2022). Determinants of the number of children born in Ethiopia, evidenced from 2019 miniEDHS: Using zero-truncated count regression models.
  • Yesilova, A., Kaydan, M. B., & Kaya, Y. (2010). MODELING INSECT-EGG DATA WITH EXCESS ZEROS USING ZERO-INFLATED REGRESSION MODELS. Hacettepe Journal of Mathematics and Statistics, 39(2), 273-282.

Sayma Verisi Modelleri Üzerine Bir Karşılaştırma: Konut Sayısına Etki Eden Faktörler Türkiye Örneği

Yıl 2024, Cilt: 17 Sayı: 1, 30 - 35, 20.11.2024
https://doi.org/10.58688/kujs.1467396

Öz

Yoksulluk çok boyutlu bir kavramdır. Yoksulluk göstergelerden bir tanesi hanenin sahip olduğu konut sayısıdır. Bu çalışmada, hane halkı bireyinin sahip olduğu konut sayısına etki eden faktörleri belirlemek için sayıma dayalı regresyon modelleri kullanılmıştır. Ayrıca, veriye en iyi uyum sağlayan regresyon modeli araştırılmıştır. Sayıma dayalı regresyon modellerinden en sık kullanılanlar klasik sayıma dayalı regresyon modelleri ve sıfır yığılmalı sayıma dayalı regresyon modelleridir. Ancak literatürde önerilmiş diğer bir regresyon modeli sıfır kesilmiş sayıma dayalı regresyon modelleridir. Bu modeller tüm veriyi analiz etmenin yarattığı zaman ve maliyet kaybının önüne geçmektedir. Bu nedenle, bu modeller sayıma dayalı verilerin olduğu durumlarda modellemede kullanılmak için iyi bir seçenektir. Çeşitli sayıma dayalı regresyon modelleri uygulamasını TÜİK’in yaptığı Gelir ve Yaşam Koşulları Araştırması veri setine uygulanmıştır. Çalışmada ele alınan modellerin performans değerlendirilmesi yapılmıştır. Bu değerlendirmeler için Akaike Bilgi Kriteri ve Log olabilirlik değeri kullanılmıştır. Sonuç olarak, sıfır kesilmiş negatif binom regresyon modeli gerçek veri setine en iyi uyum gösteren modeldir.

Kaynakça

  • Agresti A. (2002). Categorical data analysis (Second Edition), New Jersey: Wiley & Sons Incorporation.
  • Akaike, H. (1973). Information theory and extension of the maximum likelihood principle, Second International Symposium on Information Theory, Budapest: Akademiai Kiado, 267–281.
  • Altun, E. (2018). A new zero-inflated regression model with application. İstatistikçiler Dergisi: İstatistik ve Aktüerya, 11(2), 73-80.
  • Alwani, Z. Z., Ibrahim, A. I. N., Yunus, R. M., & Yusof, F. (2021). Application of zero-truncated count data regression models to air-pollution disease. In Journal of Physics: Conference Series (Vol. 1988, No. 1, p. 012096). IOP Publishing.
  • Cameron, A. C., and Trivedi, P. K. (2013). Regression analysis of count data (Second edition). New York: Cambridge university press, 128-132.
  • Durmuş, B., & Güneri, Ö. İ. (2020). An application of the generalized poisson model for over dispersion data on the number of strikes between 1984 and 2017. Alphanumeric Journal, 8(2), 249-260.
  • Gevrekci, Y., Guneri, O. I., Takma, C., & Yesilova, A. (2022). Comparison of different count models for investigation of some environmental factors affecting stillbirth in holsteins. Indian Journal of Animal Research, 56(9), 1158-1163.
  • Greene, W. H., (1994). Accounting for excess zeros and sample selection in poisson and negative binomial regression models. New York University Working Paper, 94(10), 1- 37.
  • Grogger, J. T., and Carson, R. T. (1991). Models for truncated counts. Journal of applied econometrics, 6(3), 225-238.
  • Hilbe, J. M. (2014). Modelling Count Data. New York: Cambridge University Press, 20- 170.
  • İnternet: https://data.tuik.gov.tr/Bulten/Index?p=Gelir-ve-Yasam-Kosullari-Arastirmasi2018-30755. Son erişim tarihi: 25.09.2020.
  • İşçi Güneri, Ö., & Durmuş, B. (2021). Models for Overdispersion Count Data with Generalized Distribution: An Application to Parasites Intensity. Journal of New Theory(35), 48-61. https://doi.org/10.53570/jnt.902066
  • Kim, D. W., Deo, R. C., Park, S. J., Lee, J. S., and Lee, W. S. (2019). Weekly heat wave death prediction model using zero-inflated regression approach. Theoretical and Applied Climatology, 137(1-2), 823-838.
  • Lawal, B. H. (2022). Zero-Truncated Models applied to the Nigerian National Health Insurance Data. BENIN JOURNAL OF STATISTICS , Vol. 5, pp. 1– 20.
  • Min, Y., and Agresti, A. (2005). Random effect models for repeated measures of zeroinflated count data. Statistical modelling, 5(1), 1-19.
  • Tüzen, M., F. and Erbaş, S. (2017). A comparison of count data models with an application to daily cigarette consumption of young persons. Communıcatıons In Statistics Theory And Methods, 47(23), 5825-5844.
  • Worku, G., Tadesse, G., Arega, A., & Tesfaw, D. (2022). Determinants of the number of children born in Ethiopia, evidenced from 2019 miniEDHS: Using zero-truncated count regression models.
  • Yesilova, A., Kaydan, M. B., & Kaya, Y. (2010). MODELING INSECT-EGG DATA WITH EXCESS ZEROS USING ZERO-INFLATED REGRESSION MODELS. Hacettepe Journal of Mathematics and Statistics, 39(2), 273-282.
Toplam 18 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular Uygulamalı İstatistik
Bölüm Makaleler
Yazarlar

Onur Şentürk 0000-0002-6752-4963

Hülya Olmuş 0000-0002-8983-708X

Yayımlanma Tarihi 20 Kasım 2024
Gönderilme Tarihi 10 Nisan 2024
Kabul Tarihi 11 Eylül 2024
Yayımlandığı Sayı Yıl 2024 Cilt: 17 Sayı: 1

Kaynak Göster

APA Şentürk, O., & Olmuş, H. (2024). Sayma Verisi Modelleri Üzerine Bir Karşılaştırma: Konut Sayısına Etki Eden Faktörler Türkiye Örneği. Kafkas Üniversitesi Fen Bilimleri Enstitüsü Dergisi, 17(1), 30-35. https://doi.org/10.58688/kujs.1467396