Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions
Yıl: 2021 Cilt: 13 Sayı: 3 Sayfa Aralığı: 225 - 235 Metin Dili: İngilizce DOI: 10.5336/biostatic.2021-85803 İndeks Tarihi: 14-05-2022
Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions
Öz: Objective: Receiver operating characteristic (ROC)curve is a statistical method used to examine the actual effective ness of a diagnostic test or a biomarker in a comprehensive andreliable way. Several methods have been proposed to estimate ROCcurve properly. The aim of the present study is to compare recent ROC curve estimation methods for different distribution and sam ple sizes. Material and Methods: Log-concave density andsmooth log-concave density estimate based ROC curve estimation,kernel based ROC curve estimation with Gaussian, Epanechnikov,rectangular, triangular kernels, and binormal ROC estimationmethods were compared for different simulation scenarios. Re sults: The ROC curve estimation methods based on kernel esti mates gave their best performances when the biomarker values ofnon-diseased group are normal but the biomarker values of the dis eased group are right-skewed, with a notable difference from othermethods. Epanechnikov and rectangular kernel methods yieldedbetter performance than other kernel methods in small sample sizes;but this difference disappeared as the sample size increased. Themethods based on kernel or log-concave density estimate gave theirworst results for the simulation scenario where the data were non normal but symmetric. Conclusion: The performances of the othermethods examined in the study exceeded the performance of thebinormal method in highly skewed data in both groups and whenthe distribution of diseased and non-diseased populations wereright-skewed and normal, respectively.
Anahtar Kelime: Farklı Dağılımlar ve Farklı Çekirdek Fonksiyonları içinROC Eğrisi Tahmin Yöntemlerinin Performanslarının İncelenmesi
Öz: Amaç: Alıcı işletim karakteristiği [receiver operatingcharacteristic (ROC)] eğrisi, bir tanı testinin veya bir biyobelirtecingerçek etkinliğini kapsamlı ve güvenilir bir şekilde incelemek içinkullanılan istatistiksel bir yöntemdir. ROC eğrisini doğru bir şekilde tahmin etmek için çeşitli yöntemler önerilmiştir. Bu çalışmanınamacı, farklı dağılım ve örneklem büyüklükleri için güncel ROCeğrisi tahmin yöntemlerini karşılaştırmaktır. Gereç ve Yöntem ler: Log-konkav yoğunluk ve düzgün log-konkav yoğunluk tahminitabanlı ROC eğrisi tahmin yöntemi, Gaussian, Epanechnikov, dikdörtgen, üçgen kernel fonksiyonu kullanan kernel tabanlı ROC eğ risi tahmin yöntemleri ve binormal ROC eğrisi tahmin yöntemlerifarklı simülasyon senaryoları kullanılarak karşılaştırılmıştır. Bulgu lar: Kernel tahmincilerine dayanan ROC eğrisi tahmin yöntemleri,sağlıklı grubun biyobelirteç değerlerinin normal dağılım, hasta grubun biyobelirteç değerlerinin sağa çarpık dağılım gösterdiği du rumda, diğer yöntemlerden büyük farkla en iyi performansı göster miştir. Epanechnikov vedikdörtgen kernel yöntemleri, diğer kernelyöntemlerinden küçük örneklemlerde daha iyi performans göster mekle birlikte aralarındaki fark, örneklem büyüklüğündeki artışlaortadan kalkmıştır. Kernel ve log-konkav yoğunluk tahminine daya lı yöntemler, verinin normal olmadığı fakat simetrik olduğu durum da en kötü sonucu vermişlerdir. Sonuç: Çalışmada incelenen yön temlerin performansları, her iki grupta yüksek oranda çarpık verilerolması durumunda ve hasta ve sağlıklı popülasyonların dağılımlarısırasıyla sağa çarpık dağılım ve normal dağılım olduğunda,binormal yöntemin performansını geçmiştir.
Anahtar Kelime: Belge Türü: Makale Makale Türü: Araştırma Makalesi Erişim Türü: Erişime Açık
- 1. Swets JA, Pickett RM. Evaluation of Diagnostic Systems: Methods from Signal Detection Theory. 1st ed. New York: Academic Press Inc.; 1982. [Link]
- 2. Hanley JA. Receiver operating characteristic (ROC) methodology: the state of the art. Crit Rev Diagn Imaging. 1989;29(3):307-35. [PubMed]
- 3. Begg CB. Advances in statistical methodology for diagnostic medicine in the 1980's. Stat Med. 1991;10(12):1887-95. [Crossref] [PubMed]
- 4. Zhou XH, Obuchowski N, McClish D. Statistical Methods in Diagnostic Medicine. 2nd ed. New Jersey: John Wiley & Sons Inc; 2011. [Crossref]
- 5. Hsieh F, Turnbull B. Nonparametric and semiparametric estimation of the receiver operating characteristic curve. Annals of Statistics. 1996;24(1):25- 40. [Crossref]
- 6. Pepe M, Longton G, Janes H. Estimation and comparison of receiver operating characteristic curves. Stata J. 2009;9(1):1. [Crossref] [PubMed] [PMC]
- 7. Zou KH, Hall WJ, Shapiro DE. Smooth non-parametric receiver operating characteristic (ROC) curves for continuous diagnostic tests. Stat Med. 1997;16(19):2143-56. [Crossref] [PubMed]
- 8. Jokiel-Rokita A, Pulit M. Nonparametric estimation of the ROC curve based on smoothed empirical distribution functions. Stat Comput. 2013;23(6):703- 12. [Crossref]
- 9. Zou KH, Hall WJ, Shapiro DE. Smooth non-parametric receiver operating characteristic (ROC) curves for continuous diagnostic tests. Stat Med. 1997;16(19):2143-56. [Crossref] [PubMed]
- 10. Lloyd CJ. Using smoothed receiver operating characteristic curves to summarize and compare diagnostic systems. JASA. 1998;93(444):1356- 64. [Crossref]
- 11. Lloyd CJ, Yong Z. Kernel estimators of the ROC curve are better than empirical. Statistics and Prob Letters. 1999;44(3):221-8. [Crossref]
- 12. Rufibach K. A smooth ROC curve estimator based on log-concave density estimates. Int J Biostat. 2012;8(1). [Crossref] [PubMed]
- 13. Rufibach K. Computing maximum likelihood estimators of a logconcave density function. J Statist Comp Sim. 2007;77(7):561-74. [Crossref]
- 14. Zhou XH, Lin H. Semi-parametric maximum likelihood estimates for ROC curves of continuous-scale tests. Stat Med. 2008;27(25):5271-90. [Crossref] [PubMed] [PMC] 1 5. Wang D, Cai X. Smooth ROC curve estimation via Bernstein polynomials. PLoS One. 2021;16(5):e0251959. [Crossref] [PubMed] [PMC]
- 16. Sheather SJ, Jones MC. A reliable data-based bandwidth selection method for kernel density estimation. J R Statist Soc B. 1991;53(3):683- 90. [Crossref]
- 17. Robin X, Turck N, Hainard A, Tiberti N, Lisacek F, Sanchez JC, et al. pROC: an open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinformatics. 2011;12:77. [Crossref] [PubMed] [PMC]
| APA | Sigirli D (2021). Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. , 225 - 235. 10.5336/biostatic.2021-85803 |
| Chicago | Sigirli Deniz Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. (2021): 225 - 235. 10.5336/biostatic.2021-85803 |
| MLA | Sigirli Deniz Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. , 2021, ss.225 - 235. 10.5336/biostatic.2021-85803 |
| AMA | Sigirli D Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. . 2021; 225 - 235. 10.5336/biostatic.2021-85803 |
| Vancouver | Sigirli D Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. . 2021; 225 - 235. 10.5336/biostatic.2021-85803 |
| IEEE | Sigirli D "Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions." , ss.225 - 235, 2021. 10.5336/biostatic.2021-85803 |
| ISNAD | Sigirli, Deniz. "Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions". (2021), 225-235. https://doi.org/10.5336/biostatic.2021-85803 |
| APA | Sigirli D (2021). Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. Türkiye Klinikleri Biyoistatistik Dergisi, 13(3), 225 - 235. 10.5336/biostatic.2021-85803 |
| Chicago | Sigirli Deniz Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. Türkiye Klinikleri Biyoistatistik Dergisi 13, no.3 (2021): 225 - 235. 10.5336/biostatic.2021-85803 |
| MLA | Sigirli Deniz Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. Türkiye Klinikleri Biyoistatistik Dergisi, vol.13, no.3, 2021, ss.225 - 235. 10.5336/biostatic.2021-85803 |
| AMA | Sigirli D Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. Türkiye Klinikleri Biyoistatistik Dergisi. 2021; 13(3): 225 - 235. 10.5336/biostatic.2021-85803 |
| Vancouver | Sigirli D Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions. Türkiye Klinikleri Biyoistatistik Dergisi. 2021; 13(3): 225 - 235. 10.5336/biostatic.2021-85803 |
| IEEE | Sigirli D "Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions." Türkiye Klinikleri Biyoistatistik Dergisi, 13, ss.225 - 235, 2021. 10.5336/biostatic.2021-85803 |
| ISNAD | Sigirli, Deniz. "Evaluating the Performances of ROC Curve EstimationMethods for Different Distributions andDifferent Kernel Functions". Türkiye Klinikleri Biyoistatistik Dergisi 13/3 (2021), 225-235. https://doi.org/10.5336/biostatic.2021-85803 |
