Accumetric Test Performance Analysis software, receiver operating characteristic analysis, nonparametric, sensitivity, specificity, likelihood ratio, predictive value, kappa, weighted kappa, intraclass correlation

Measures of test performance included in Accumetric Test Performance Analysis are:

    Measures of reliability, with most including standard errors and precise 95%
    confidence intervals:

    * Kappa, with standard error and 95% confidence interval, and comparison of
      two kappas or of a kappa with a fixed value
    * Weighted kappa, with standard error and 95% confidence interval, and
       comparison of two weighted kappas or of a weighted kappa with a fixed value
    * Intraclass correlation

    Measures of validity, including standard errors and precise 95% confidence
    intervals:

    * Counts of true and false positives and negatives
    * Sensitivity, with standard error and 95% confidence interval
    * Specificity, with standard error and 95% confidence interval
    * False positive and false negative proportions, with standard error and 95%
      confidence interval
    * Correct classification and misclassification proportions, with standard error
       and 95% confidence interval
    * Likelihood ratios of a positive and negative test and stratum-specific likelihood
       ratios, with standard error and 95% confidence interval
    * Positive and negative predictive values at any pre-test probability of disease,
       with standard error and 95% confidence interval
    * Post-test probabilities of disease at any pre-test probability of disease, with
       standard error and 95% confidence interval
    * Nonparametric receiver operating characteristic analysis, with
      area under the curve, its standard error and its 95% confidence interval, and
      statistical comparison of two ROC curves from independent samples or of two
      or three curves from identical samples

       More on ROC analysis