International Journal of Statistics and Systems

  • Year: 2010
  • Volume: 5
  • Issue: 2

On the Relationship between the Odds Ratio and the Area under the ROC Curve in the context of Logistic Regression for Comparing Several Biomarkers

  • Author:
  • R. Vishnu Vardhan1, K.V.S. Sarma2
  • Total Page Count: 8
  • DOI:
  • Page Number: 165 to 172

1Center for Mathematical Sciences - DST, C.R.Rao Advanced Institute of Mathematics Statistics and Computer Science (AIMSCS), UoH Campus, Hyderabad-500 046.

2Department of Statistics, Sri Venkateswara Univeristy, Tirupati-517 502.

null

Abstract

While comparing effectiveness of several biomarkers as determinants of a disease, it is observed that the obtained area under the receiver operating characteristic (ROC) curve has a similar behavior like that of the odds ratio (OR) obtained from binary logistic regression. For a biomarker if the OR is high (low), the area under the curve (AUC) of ROC is also high (low). This paper reports some empirical results on the functional relationship between AUC and OR by using simulated data. Spreadsheets-based templates are also developed to perform the calculations.

Keywords

AUC, Odds Ratio, ROC Curve and Logistic Regression