Gamma Regression for the Poverty Rate in Türkiye

Year 2025, Issue: 42, 22 - 35, 25.06.2025

Müge Borazan , Serpil Aktaş

https://doi.org/10.26650/ekoist.2025.42.1523478

Abstract

The family of gamma regression models makes the assumption that a linear predictor with unknown coefficients and a link function, such as an identity, inverse, or logarithmic function, are related to a set of predictors by means of the mean of the dependent variable with a gamma distribution. This model also has a shape parameter, which can either depend on a collection of regressors via a link function or be constant like a logarithm function. The analysis of positive random variables can make extensive use of the gamma distribution. When the dependent variable has a real value between 0 and ∞, gamma regression makes sense. In this study, a gamma regression model is used to examine the relationship between poverty rates and household education levels in Türkiye for the period 2006–2023. The Turkish Statistical Institute (TurkStat) provided the study data set. According to TurkStat’s definition of poverty, which is 50% of the comparable household disposable median income, the poverty rates in the data set indicate the percentage of people who are at danger of becoming impoverished. The Gamma Regression Model reveals that the poverty rate is significantly influenced by education level. Furthermore, there are notable variations in poverty rates among education levels, according to the post hoc anlayses we performed to compare poverty rates across educational levels.

Keywords

Gamma regression, parameter estimation, poverty, poverty rate

References

• Algamal, Z. Y. (2018). Developing a ridge estimator for the gamma regression model, Journal of Chemometrics, vol. 32, no. 10, p. e 3054. google scholar
• Amin, M., Qasim, M. and Amanullah, M. (2019). Performance of Asar and Genç and Huang and Yang’s two-parameter estimation methods for the gamma regression model, Iranian Journal of Science and Technology, Transactions A: Science, vol. 43, no. 6, pp. 2951–2963. google scholar
• A. M. Al-Abood and D. H. Young, (1986). Improved deviance goodness of fit statistics for a gamma regression model, Communications in Statistics Theory and Methods, vol. 15, no. 6, pp. 1865–1874. google scholar
• Barry R. Chiswick and Michael D. McCarthy, (1977). A Note on Predicting the Poverty Rate. google scholar
• Basu, A. and Manning, W.G. (2006). A test for proportional hazards assumption within the class of exponential conditional mean models, Health Serv. Outcomes Res. Methodol. 6 (2006), pp. 81–100. google scholar
• Blumenstock, J., Cadamuro, G., & On, R. (2015). Predicting poverty and wealth from mobile phone metadata. Science, 350(6264), 1073– 1076. https://doi.org/10.1126/science.aac4420. google scholar
• Bossio, M. and Cuervo, E.C., (2015). Gamma regression models with the Gammareg R package, Communicaciones en Estad´ıstica Diciembre, Vol. 8, No. 2, pp. 211–223. google scholar
• Chen, X., Aravkin, A.Y., Martin, R.D.(2018). Generalized Linear Model for Gamma Distributed Variables via Elastic Net Regularization, arXiv:1804.07780v1 [stat.ME], https://doi.org/10.48550/arXiv.1804.07780. google scholar
• Chiswick, B. R., & McCarthy, M. D. (1977). A note on predicting the poverty rate. The Journal of Human Resources, 12(3), 396-400. google scholar
• Elbers, C., J. O. Lanjouw, and P. Lanjouw. (2003). Micro Level Estimation of Poverty and Inequality, Econometrica, 71, 355–64. google scholar
• Economic Commission for Latin America and the Caribbean (ECLAC), (2019). Income poverty measurement: updated methodology and results, ECLAC Methodologies, No. 2 (LC/PUB.2018/22-P), Santiago. google scholar
• Dobson, A. and Barnett, A. (2008). An Introduction to Generalized Linear Models, 3rd edn. Boca Raton, FL: Chapman and Hall. google scholar
• Dunder E., Gumustekin, S. & Cengiz, M.A. (2016): Variable selection in gamma regression models via artificial bee colony algorithm, Journal of Applied Statistics, DOI: 10.1080/02664763.2016.1254730. google scholar
• Fox, J. (2008). Applied Regression Analysis and Generalized Linear Models, 2nd edn. Thousand Oaks, CA: Sage. google scholar
• Gallucci, M. (2019). GAMLj: General analyses for linear models. [jamovi module]. Retrieved from https://gamlj.github.io/. google scholar
• Gill, J. (2001). Generalized Linear Models: A Unified Approach. Thousand Oaks, CA: Sage. google scholar
• Hardin, J.W., Hilbe, J.M. (2007) Generalized linear models and extensions. 2nd ed. College Station, TX: Stata Press. google scholar
• Hoogstra, B., Velichety, S., Zhang, C., (2024): Developing a contextual model of poverty prediction using data science and analytics – The case of Shelby County, Decision Support Systems, Volume 177. google scholar
• Jean, N., Burke, M., Xie, M., Davis, W. M., Lobell, D. B., & Ermon, S. (2016). Combining satellite imagery and machine learning to predict poverty. Science, 353(6301), 790–794. https://doi. org/10.1126/science. aaf7894. google scholar
• Liang, K. and Zeger, S. (1986). Longitudinal data analysis using generalized linear models. Biometrika 73(1), 13–22. google scholar
• Lindsey, J. (1997). Applying Generalized Linear Models. New York: Springer. google scholar
• Mathiassen, A. (2013). Testing Prediction Performance of Poverty Models: Emprical Evidence from Uganda. Review of Income and Wealth Series 59, Number 1, March 2013, DOI: 10.1111/roiw.12007. google scholar
• Mathiassen, A., (2009). A Model Based Approach for Predicting Annual Poverty Rates Without Expenditure Data, Journal of Economic Inequality, 7, 117–35. google scholar
• M. Wasef Hattab, (2016). A derivation of prediction intervals for gamma regression, Journal of Statistical Computation and Simulation, vol. 86, no. 17, pp. 3512–3526. google scholar
• McCullagh, P. and Nelder, J.A. (1989) Generalized Linear Models. 2nd Edition, Chapman and Hall, London.http://dx.doi.org/10.1007/978-1-4899-3242-6. google scholar
• Manning, W.G. and Mullahy, J. (2001). Estimating log models: To transform or not to transform? J. Health Econ. 20 (2001), pp. 461–494. google scholar
• McLeod, A.I. and Xu, C. (2010). bestglm: Best subset GLM. Available at http://CRAN.Rproject.org/packag8e = bestglm. google scholar
• Myers, R.H., Montgomery, D.C., Vining, G.G., Robinson, T.J. (2010) Generalized linear models with applications in engineering and the sciences. 2nd ed. New York: Wiley. google scholar
• Nelder, J. and Wedderburn, R. (1972). Generalized linear models. Journal of the Royal Statistical Society, Series A 132, 370–384. google scholar
• Puttanapong N., Martinez Jr.,A.M., Addawe,M., Bulan,B., Durante,R.L. and Martillan, M. (2020): Predicting Poverty Using Geospatial Data in Thailand. ADB Economics Working Paper Series, Asian Development Bank. google scholar
• Pokhriyala, N. and Jacques, D.C. (2017) Combining disparate data sources for improved poverty prediction and mapping. www.pnas.org/ cgi/doi/10.1073/pnas.1700319114. google scholar
• R Core Team (2022). R: A Language and environment for statistical computing. (Version 4.1) [Computer software]. Retrieved from https:// cran.r-project.org. (R packages retrieved from CRAN snapshot 2023-04-07). google scholar
• Usmanova, A.; Aziz, A.; Rakhmonov, D.; Osamy, W. (2022): Utilities of Artificial Intelligence in Poverty Prediction: A Review. Sustainability, google scholar
• 14, 14238. https://doi.org/ 10.3390/su142114238. google scholar
• Wijaya, D.R., Satyaning Pradnya Paramita, N.L.P, Uluwiyah, A., Rheza, M., Zahara, A., Puspita, D.R. (2020). Estimating city‑level poverty rate based on e‑commerce data with machine learning. Electronic Commerce Research (2022) 22:195–221. google scholar
• Wedderburn, R. (1974). Quasi-likelihood functions, generalized linear models, and the Gauss–Newton method. Biometrika 61(3), 439–447. google scholar
• The Journal of Human Resources, Summer, 1977, Vol. 12, No. 3 (Summer, 1977), pp. 396-400. google scholar
• The jamovi project (2023). jamovi. (Version 2.4) [Computer Software]. Retrieved from https://www.jamovi.org. google scholar