Determination of The Best-Fit Copula Family for Air Temperature Data of Ankara
Ferhan Baş Kaman1*, Hülya Olmuş2
1Department of Banking and Finance, Graduate School of Social Sciences, Yıldırım Beyazıt University , Ankara, Turkey
2Department of Statistics, Graduate School of Natural and Applied Sciences, Gazi University , Ankara, Turkey
* Corresponding author: basferhan@gmail.com
Presented at the Ist International Symposium on Innovative Approaches in Scientific Studies (ISAS 2018), Kemer-Antalya, Turkey, Apr 11, 2018
SETSCI Conference Proceedings, 2018, 2, Page (s): 138-138 , https://doi.org/
Published Date: 23 June 2018 |
973 
7
Abstract
In order to investigate the relationship between variables, it is necessary to reveal the dependency structure between variables and because of this reason, the concept of “copula” is being used in statistics literature in recent years. The main purpose of the copula function is to find out the multivariate distribution best suited for the observed data also by revealing the dependency structure. Copulas are used in many studies such as statistics, economics, actuary, financial and risk management, time series and modelling. In this study, an application of meteorological data with some copulas is given. These copula families are Gumbel, Clayton, Frank and Gaussian. The best-fit copula family is determined for the dependence structure of monthly lowest and highest air temperature records from 1985 to 2016 in Ankara. Akaike Information Criteria (AIC) and Bayesian Information Criteria (BIC) are used to determine the best-fit copula. According to the analysis results, the best-fit copula family is determined as Frank copula. In addition, the best-fit copula family is examined with scatter plots of the data obtained from copulas.
Keywords - copula, dependency structure, meteorological data, air temperature
References
Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control,
19(6), 716-723.
Nelsen, R. B. (2006). An Introduction to Copulas.(Second edition), New York:Springer, 7-211.
Schwarz, G. (1978). Estimating the dimension of a model. The annals of Statistics, 6(2), 461-464.
Sklar, A. (1959). Fonctions de repartitiona n dimesions et leursmarges. Publ. Inst. Statist. Univ., Paris 8: 229-
231.