Analyzing dependence structure of thyroid hormones: a copula approach

To model the level and structure of the dependence between thyroid hormones by using the copula approach. Materials and methods: Dependence models were constructed with the help of copula functions explaining the relationships between thyroid hormones for the data supplied by Dokuz Eylül University's Faculty of Medicine's endocrine laboratory. For this purpose, 684 patients aged 0-85 were examined. Results: Results indicated that none of the pairs of thyroid hormones exhibited a tail dependence structure; however, valid models exhibited a symmetric dependence structure. The findings implied that both the T3 and T4 levels had a significant dependence structure with TSH levels. Furthermore, Gaussian and t-copula structures were appropriate for the pairs. Conclusion: Findings were compared with the results of conventional scalar measures to establish the importance of using copula models in analyzing thyroid hormone levels. Findings showed that the copula models revealed better indicators for health scientists for more accurate dependence modeling.

Anahtar Kelimeler:

Key words: Thyroid hormones, copula, IFM, correlation, dependence

Analyzing dependence structure of thyroid hormones: a copula approach

Keywords:

Key words: Thyroid hormones, copula, IFM, correlation, dependence,

PDF

___

Ulutagay G, Nasibov E. A fuzzy inference system for thyroid disease diagnosis. Proceedings of 1st International Symposium on Computing in Science & Engineering, Gediz Üniversitesi, İzmir, 2010.
Mari D, Kotz S. Correlation and dependence. Imperial College Press; 2001.
Nelsen RB. An introduction to copulas. Springer; 1998.
Schweizer B, Wolff E. On parametric measures of dependence for random variables. Annals of Statistics 1981; 9: 879-85. 5. Joe H. Multivariate models and dependence concepts. London: Chapman and Hall; 1997. 6. Genest C, Rivest L. Statistical inference procedures for bivariate Archimedean copulas. Journal of the American Statistical Association 1993; 88: 1034-43.
Genest C, Ghoudi K, Rivest L. A semiparametric estimation procedure of dependence parameters in multivariate families of distributions. Biometrika 1995; 82: 543-552.
Oruç Ege Ö, Üçer B. A new method for local dependence map and its applications. Türkiye Klinikleri J Biosta 2009; 1: 1-8.
MathWorks Inc. MATLAB: the language of technical computing. Natick (MA): MathWorks; 2007.