Abstract
The volume under a surface (VUS) is an effective measure for evaluating the discriminating power of a diagnostic test with three ordinal diagnostic groups. In this paper, we investigate the difference of two correlated VUS’s to compare two treatments for discrimination of three-class classification data. A jackknife empirical likelihood (JEL) procedure is employed to avoid the variance estimation in the existing methods. We prove that the limiting distribution of the empirical log-likelihood ratio statistic follows a \(\chi ^2\) distribution. Extensive numerical studies show that the JEL confidence intervals outperform those based on the normal approximation method. The proposed method is also applied to the Alzheimer’s disease data.
Similar content being viewed by others
References
Arvesen, J. (1969). Jackknifing U-statistics. The Annals of Mathematical Statistics, 40(6), 2096–2100.
Balakrishnan, K. (1996). Exponential distribution: Theory, methods and applications. Amsterdam: CRC Press. illustrated edition.
Gong, Y., Peng, L., Qi, Y. (2010). Smoothed jackknife empirical likelihood method for ROC curve. Journal of Multivariate Analysis, 101, 1520–1531.
Jing, B. Y., Yuan, J., Zhou, W. (2009). Jackknife empirical likelihood. Journal of the American Statistical Association, 104, 1224–1232.
Kowalski, J., Tu, X. M. (2007). Modern applied U-statistics. Hoboken, New Jersey: Wiley.
Marshall, A. W., Olkin, I. (1967). A multivariate exponential distribution. Journal of American Statistical Association, 62, 30–41.
Mossman, D. (1999). Three-way ROC’s. Medical Decision Making, 19(1), 78–89.
Owen, A. B. (1988). Empirical likelihood ratio confidence intervals for a singlep functional. Biometrika, 75(2), 237–249.
Owen, A. B. (1990). Empirical likelihood confidence regions. The Annals of Statistics, 18, 90–120.
Pan, G., Wang, X., Zhou, W. (2013). Nonparametric statistical inference for P(X \(<\) Y \(<\) Z). The Indian Journal of Statistics, 75-A, 118–138.
Sen, P. K. (1960). On some convergence properties of U-statistics. Calcutta Statistical Association Bulletin, 10, 1–18.
Storandt, M., Hill, R. D. (1989). Very mild senile dementia of the Alzheimer type. II. Psychometric test performance. Archives of Neurology, 46(4), 383–386.
Tian, L., Xiong, C., Lai, C. Y., Vexler, A. (2011). Exact confidence interval estimation for the difference in diagnostic accuracy with three ordinal diagnostic groups. Journal of Statistical Planning and Inference, 141(1), 549–558.
Wan, S. (2012). An empirical likelihood confidence interval for the volume under ROC surface. Statistics and Probability Letters, 82(7), 1463–1467.
Wang, X. (2010). Empirical likelihood with applications. Thesis, National University of Singapore.
Wechsler, D. (1955). Manual for the Wechsler adult intelligence scale (Vol. vi). Oxford: Psychological Corporation.
Xiong, C., van Belle, G., Miller, J. P., Morris, J. C. (2006). Measuring and estimating diagnostic accuracy when there are three ordinal diagnostic groups. Statistics in Medicine, 2006(25), 1251–1273.
Yang, H., Zhao, Y. (2013). Smoothed jackknife empirical likelihood inference for the difference of ROC curves. Journal of Multivariate Analysis, 115, 270–284.
Yang, H., Zhao, Y. (2015). Smoothed jackknife empirical likelihood inference for ROC curves with missing data. Journal of Multivariate Analysis, 140, 123–138.
Acknowledgements
The authors would like to thank two reviewers and the associate editor for their helpful comments, which lead to a significant improvement in the paper. Yichuan Zhao appreciates the support from the NSA Grant (H98230-12-1-0209) and NSF Grants (DMS-1613176).
Author information
Authors and Affiliations
Corresponding author
Appendix: Proof of Theorem 1
Appendix: Proof of Theorem 1
The variance \(Var(U_n)\) can be estimated by a consistent estimator \({\hat{\sigma }^2}\) as in Sen (1960) and Arvesen (1969),
Lemma 1
We have the following conclusions.
-
(a)
The U-statistic \(U_n\buildrel a.s. \over \rightarrow \theta _0\) as \(min(n_1, n_2, n_3)\rightarrow \infty \);
-
(b)
Suppose that \(\sigma ^2_{1,0,0}>0\), \(\sigma ^2_{0,1,0}>0\), \(\sigma ^2_{0,0,1}>0\), and denote \(S^2_{n_1, n_2, n_3}=\sigma ^2_{1,0,0}/n_1+\sigma ^2_{0,1,0}/n_2+\sigma ^2_{0,0,1}/n_3\). As \(\quad min(n_1, n_2, n_3)\rightarrow \infty ,\)
$$\begin{aligned} \dfrac{U_n-\theta _0}{S_{n_1, n_2, n_3}}\buildrel d \over \rightarrow N(0,1),\quad \end{aligned}$$(3)and
$$\begin{aligned} \hat{\sigma }^2-S_{n_1, n_2, n_3}^2=o_p((min(n_1, n_2, n_3))^{-1}). \end{aligned}$$(4)
The proof of part (a) and Eqs. (3) and (4) can be found in Arvesen (1969) and Kowalski and Tu (2007).
Lemma 2
Let \(S_n=n^{-1}\sum \nolimits _{l=1}^n(\hat{V}_l-E\hat{V}_l)^2\). We assume the same conditions as (a) and (b) in Theorem 1. Then as \(n_1\rightarrow \infty \),
Proof of Lemma 2
For \(1\le l\le n_1\), it is clear that
and
As Pan et al. (2013) and Wang (2010) did, we have that
For \(n_1< l\le n_1+n_2\), one has that
For \(n_1+n_2< l\le n\), we have that (see Pan et al. 2013)
Therefore,
From the LLN of U-statistic, we have the conclusion \(U_n-\theta _0=O_p(n_1^{-1/2})\). The second term in Eq. (5) is
Moreover, the 1st term of Eq. (5) is (cf. Wang 2010)
Using Eq. (4), we prove Lemma 2. \(\square \)
Lemma 3
Let \(Q_n=\displaystyle \max _{1\le l\le n}|\hat{V}_l-\theta _0|\). Under the same conditions as in Lemma 2, we have \(Q_n=o_p(n^{1/2})\) and \(n^{-1}\sum \nolimits _{l=1}^n|\hat{V}_l-\theta _0|^3=o_p(n^{1/2})\).
Proof of Lemma 3
For \(1\le l\le n_1\), we have (see Wang 2010)
Note that \(|V_{l,0,0}|\le \tilde{H}_n\), and \(|U_n|\le \tilde{H}_n\), where
Therefore, \(|\hat{V}_l-E\hat{V}_l| \le c^*\tilde{H}_n+c^*\tilde{H}_n+ c^{*}|U_n-\theta |\), where \(c^*\) is a constant. Similar to Pan et al. (2013) and Wang (2010), we have \(\tilde{H}_n=o_p(n^{1/2})\) and \(U_n-\theta _0 =O_p(n^{-1/2})\). Therefore, \(|\hat{V}_l-E\hat{V}_l|=o_p(n^{1/2})\) for \(1\le l \le n_1\). For \(n_1<l\le n_1+n_2\), \(|\hat{V}_l-E\hat{V}_l| \le 2c^*\tilde{H}_n+c^*|U_n-\theta _0|.\) Thus, \(|\hat{V}_l-E\hat{V}_l|=o_p(n^{1/2})\) for \(n_1<l\le n_1+n_2\). For \(n_1+n_2<l\le n\), we have \(|\hat{V}_l-E\hat{V}_l|=o_p(n^{1/2})\) similarly as before. Thus \(Q_n=o_p(n^{1/2}).\) Hence,
\(\square \)
Proof of Theorem 1
The proof follows the same lines of Pan et al. (2013). Recall \(U_n={1}/{n}\displaystyle \sum _{l=1}^n\hat{V}_l\) and \(\theta _0={1}/{n}\displaystyle \sum _{l=1}^n E \hat{V}_l\). Then
We have that \(|\gamma |=O_p(n^{-1/2})\). Using Taylor’s expansion, one has that
Let \(F_0= {1}/{n}\displaystyle \sum _{l=1}^n\dfrac{\gamma ^2(\hat{V}_l-E \hat{V}_l)^3}{1+\gamma (\hat{V}_l-E \hat{V}_l)}\). By Eq. (2), we have that
One can obtain \(F_0 =o_p(n^{-1/2})\). Thus Eq. (6) can be re-expressed as follows,
Combining Lemmas 1, 2 and 3, we finish the proof. \(\square \)
About this article
Cite this article
An, Y., Zhao, Y. Jackknife empirical likelihood for the difference of two volumes under ROC surfaces. Ann Inst Stat Math 70, 789–806 (2018). https://doi.org/10.1007/s10463-017-0631-z
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-017-0631-z