Abstract
In this paper, a general regression model with responses missing at random is considered. From an imputed rank-based objective function, a rank-based estimator is derived and its asymptotic distribution is established under mild conditions. Inference based on the normal approximation approach results in under coverage or over coverage issues. In order to address these issues, we propose an empirical likelihood approach based on the rank-based objective function, from which its asymptotic distribution is established. Extensive Monte Carlo simulation experiments under different settings of error distributions with different response probabilities are considered. The simulation results show that the proposed approach has better performance for the regression parameters compared to the normal approximation approach and its least-squares counterpart. Finally, a data example is provided to illustrate our method.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bindele, H. F. (2015). The signed-rank estimator for nonlinear regression with responses missing at random. Electronic Journal of Statistics, 9(1), 1424–1448.
Bindele, H. F., & Abebe, A. (2012). Bounded influence nonlinear signed-rank regression. Canadian Journal of Statistics, 40(1), 172–189.
Bindele, H. F., & Abebe, A. (2015). Semi-parametric rank regression with missing responses. Journal of Multivariate Analysis, 142, 117–132.
Bindele, H. F., & Zhao, Y. (2016). Signed-rank regression inference via empirical likelihood. Journal of Statistical Computation and Simulation, 86(4), 729–739.
Bindele, H. F., & Zhao, Y. (in press). Rank-based estimating equation with non-ignorable missing responses via empirical likelihood. Statistica Sinica.
Bindele, H. F. A. (2017). Strong consistency of the general rank estimator. Communications in Statistics - Theory and Methods, 46(2), 532–539.
Brunner, E., & Denker, M. (1994). Rank statistics under dependent observations and applications to factorial designs. Journal of Statistical Planning and Inference, 42(3), 353–378.
Chen, S. X., Peng, L., & Qin, Y.-L. (2009). Effects of data dimension on empirical likelihood. Biometrika, 96(3), 711–722.
Cheng, P. E. (1994). Nonparametric estimation of mean functionals with data missing at random. Journal of the American Statistical Association, 89(425), 81–87.
Delecroix, M., Hristache, M., & Patilea, V. (2006). On semiparametric estimation in single-index regression. Journal of Statistical Planning and Inference, 136(3), 730–769.
Einmahl, U., & Mason, D. M. (2005). Uniform in bandwidth consistency of kernel-type function estimators. The Annals of Statistics, 33(3), 1380–1403.
Gong, Y., Peng, L., & Qi, Y. (2010). Smoothed jackknife empirical likelihood method for ROC curve. Journal of Multivariate Analysis, 101(6), 1520–1531.
Healy, M., & Westmacott, M. (1956). Missing values in experiments analysed on automatic computers. Journal of the Royal Statistical Society. Series C (Applied Statistics), 5(3), 203–206.
Hettmansperger, T. P., & McKean, J. W. (2011). Robust Nonparametric Statistical Methods. Monographs on Statistics and Applied Probability (Vol. 119, 2nd ed.). Boca Raton, FL: CRC Press.
Hjort, N. L., McKeague, I. W., & Van Keilegom, I. (2009). Extending the scope of empirical likelihood. The Annals of Statistics, 37(3), 1079–1111.
Jaeckel, L. A. (1972). Estimating regression coefficients by minimizing the dispersion of the residuals. Annals of Mathematical Statistics, 43, 1449–1458.
Jing, B.-Y., Yuan, J., & Zhou, W. (2009). Jackknife empirical likelihood. Journal of the American Statistical Association, 104(487), 1224–1232.
Lahiri, S. N., & Mukhopadhyay, S. (2012). A penalized empirical likelihood method in high dimensions. The Annals of Statistics, 40(5), 2511–2540.
Little, R. J. (1992). Regression with missing x’s: A review. Journal of the American Statistical Association, 87(420), 1227–1237.
Little, R. J. A., & Rubin, D. B. (1987). Statistical analysis with missing data. Wiley Series in Probability and Mathematical Statistics: Applied Probability and Statistics. New York: Wiley.
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data. Wiley Series in Probability and Statistics (2nd ed.). Hoboken, NJ: Wiley-Interscience [John Wiley & Sons].
Owen, A. B. (1988). Empirical likelihood ratio confidence intervals for a single functional. Biometrika, 75(2), 237–249.
Owen, A. B. (1990). Empirical likelihood ratio confidence regions. The Annals of Statistics, 18(1), 90–120.
Owen, A. B. (2001). Empirical Likelihood. Chapman & Hall/CRC Monographs on Statistics & Applied Probability. Boca Raton, FL: CRC Press.
Qin, J., & Lawless, J. (1994). Empirical likelihood and general estimating equations. The Annals of Statistics, 22(1), 300–325.
Rubin, D. B. (1976). Inference and missing data. Biometrika, 63(3), 581–592.
Rubin, D. B. (2004). Multiple imputation for nonresponse in surveys. Wiley Classics Library (Vol. 81). Hoboken, NJ: Wiley.
Silverman, B. W. (1986). Density estimation for statistics and data analysis (Vol. 26). Boca Raton, FL: CRC Press.
Tang, C. Y., & Leng, C. (2010). Penalized high-dimensional empirical likelihood. Biometrika, 97(4), 905–920.
Van Buuren, S. (2012). Flexible imputation of missing data. Chapman & Hall/CRC Interdisciplinary Statistics. Boca Raton, FL: Taylor & Francis.
Wang, C. Y., Wang, S., Zhao, L.-P., & Ou, S.-T. (1997). Weighted semiparametric estimation in regression analysis with missing covariate data. Journal of the American Statistical Association, 92(438), 512–525.
Wang, Q., Linton, O., & Härdle, W. (2004). Semiparametric regression analysis with missing response at random. Journal of the American Statistical Association, 99(466), 334–345.
Wang, Q., & Rao, J. N. K. (2002). Empirical likelihood-based inference under imputation for missing response data. The Annals of Statistics, 30(3), 896–924.
Wang, Q., & Sun, Z. (2007). Estimation in partially linear models with missing responses at random. Journal of Multivariate Analysis, 98(7), 1470–1493.
Yang, H., & Zhao, Y. (2012). New empirical likelihood inference for linear transformation models. Journal of Statistical Planning and Inference, 142(7), 1659–1668.
Yu, W., Sun, Y., & Zheng, M. (2011). Empirical likelihood method for linear transformation models. Annals of the Institute of Statistical Mathematics, 63(2), 331–346.
Zhang, Z., & Zhao, Y. (2013). Empirical likelihood for linear transformation models with interval-censored failure time data. Journal of Multivariate Analysis, 116, 398–409.
Zhao, L. P., Lipsitz, S., & Lew, D. (1996). Regression analysis with missing covariate data using estimating equations. Biometrics, 52(4), 1165–1182.
Acknowledgements
The authors would like to thank the two reviewers for their helpful comments. The research of Yichuan Zhao is supported by the National Security Agency grant.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
This Appendix contains assumptions used in the development of theoretical results as well as the proof of the main results.
1.1 Assumptions
- (I 1 ):
-
φ is a nondecreasing, bounded, and twice continuously differentiable score function with bounded derivatives, defined on (0, 1), and, satisfying:
$$\displaystyle \begin{aligned}\quad \int_{0}^{1}\varphi(u)du=0\quad and\quad \int_{0}^{1}\varphi^{2}(u)du=1.\end{aligned}$$ - (I 2 ):
-
g(⋅) being a function of two variables x and β, it is required that g has continuous derivatives with respect to β that are bounded up to order 3 by p-integrable functions of x, independent of β, p ≥ 1.
- (I 3 ):
-
K(⋅) is a regular kernel of order r > 2, with window b n satisfying \(\displaystyle {nb_{n}^{4r}\to 0}\), \(C(\log {n}/n)^{\gamma }<b_{n}<h_{n}\), for any C > 0, γ = 1 − 2∕p, p > 2 and h n is a bandwidth such that \(C(\log {n}/n)^{\gamma }<h_{n}<1\) with h n → 0 as n →∞.
- (I 4 ):
-
supx E[|Y |p|x = x] < ∞, for p ≥ 1 and infx Δ(x) > 0.
- (I 5 ):
-
For fixed n, \({\boldsymbol \beta }_{0,n}\in Int(\mathcal {B})\) is the unique minimizer of E[D n(β)] such that limn→∞ β 0,n = β 0
- (I 6 ):
-
The model error has a distribution with a finite Fisher information.
- (I 6 ):
-
\(Var(\sqrt {n}S_{n}^{j}({\boldsymbol \beta }_0))\to \varSigma _{{\boldsymbol \beta }_{0}}^{j}\), where \(\varSigma _{{\boldsymbol \beta }_{0}}^{j}\) is positive definite.
- (I 7 ):
-
Set H = B(B τ B)−1 B τ, where B = ∇β g(x, β 0). H is the projection matrix onto the column space of B, which in this case represents the tangent space generated by B and let h iin, i = 1, ⋯ , n, be the leverage values that stand for the diagonal entries of H. We assume that limn→∞max1≤i≤n h iin = 0
Assumptions (I 2) − (I 4) are necessary and sufficient to ensure the strong consistency of \(\widehat {\pi }({\mathbf x})\) used in the imputation process. On the other hand, assumptions (I 1), (I 5) − (I 7) together with the previous assumptions are necessary to establish the asymptotic properties (consistency and asymptotic normality distribution) of the rank-based estimators of β 0. An elaborate discussion about these assumptions can be found in Hettmansperger and McKean (2011), Bindele and Abebe (2012), and Bindele (2017).
By definition of \(S_{n}^{j}({\boldsymbol \beta })\), \(\widehat {{\boldsymbol \beta }}_{n}^{j}\) is solution to the equation \(S_{n}^{j}({\boldsymbol \beta })=\mathbf {0}\). As in Brunner and Denker (1994), assume without loss of generality that ∥λ i∥ = 1 and define
The following lemma due to Brunner and Denker (1994) is a key for establishing asymptotic normality of the rank gradient function for dependent data.
Lemma 3.1
Let ς jn be the minimum eigenvalue of W jn = V ar(U jn) with U jn given by
Suppose that ς jn ≥ Cn a for some constants \(C, a \in \mathbb {R}\) and m(n) is such that M 0 n γ ≤ m(n) ≤ M 1 n γ for some constants 0 < M 0 ≤ M 1 < ∞ and 0 < γ < (a + 1)∕2. Then \(\displaystyle {m(n){\mathbf W}_{jn}^{-1}T_{n}^{j}({\boldsymbol \beta }_{0})}\) is asymptotically standard multivariate normal, provided φ is twice continuously differentiable with bounded second derivative.
We provide a sketch of the proof of this lemma. A detailed proof can be found in Brunner and Denker (1994).
Proof
Set
Brunner and Denker (1994) showed that W jn = n 2 V ar(B jn), as U n = nB n. From its definition, \(S_{n}^{j}({\boldsymbol \beta })\) can be rewritten as
By (I 5), since \({\boldsymbol \beta }_{0}=\displaystyle \lim _{n\to \infty } \operatorname *{\mathrm {Argmin}}_{{\boldsymbol \beta }\in \mathcal {B}}E\{D_{n}^{j}({\boldsymbol \beta })\}\), we have \(E\{S_{n}^{j}({\boldsymbol \beta }_{0})\}\to \mathbf {0}\) as n →∞. From the fact that Var(ε|x) > 0, there exists a positive constant C such that ς jn ≥ Cn 2 which satisfies the assumptions of Lemma 3.1, as φ is twice continuously differentiable with bounded derivatives, γ < (a + 1)∕2 with a = 2, M 0 = M 1 = 1, γ = 1, and m(n) = n. Thus, for n large enough, \(\displaystyle {n{\mathbf W}_{jn}^{-1}T_{n}^{j}({\boldsymbol \beta }_{0})\approx n{\mathbf W}_{jn}^{-1}S_{n}^{j}({\boldsymbol \beta }_{0})}\), which converges to a multivariate standard normal, by Lemma 3.1. A direct application of Slutsky’s Lemma and putting Σ jn = n −1∕2 W jn give \(\displaystyle \sqrt {n}S_{n}^{j}({\boldsymbol \beta }_{0}) \ \xrightarrow {\mathcal {D}} \ N_{p}(\mathbf {0}, \varSigma _{{\boldsymbol \beta }_{0}}^{j})\), j = 1, 2, where \( \varSigma _{{\boldsymbol \beta }_{0}}^{j}=\displaystyle {\lim _{n\to \infty }\varSigma _{jn}\varSigma _{jn}^{\tau }}\).
Proof of Theorem 3.2. Let C be an arbitrary positive constant. Recall from Eq. (3.5) that the log likelihood ratio of β 0 is given by
Under (I 1) and (I 2), there exist a positive constant M and a function h ∈ L p, p ≥ 1 such that |φ(t)|≤ M for all t ∈ (0, 1), and ∥∇β g(x i, β 0)∥≤ h(x i), where ∥⋅∥ stands for the L 2- norm. From this, max1≤i≤n∥∇β g(x i, β 0)∥ = o p(n 1∕2) since E(|h(x i)|P) < ∞, p ≥ 1. Also, since \(\varSigma _{jn}\varSigma _{jn}^{\tau }\to \varSigma _{{\boldsymbol \beta }_{0}}^{j}\;\; a.s.\), Σ jn is almost surely bounded. Thus, ∥η ij(β 0)∥≤ M ×max1≤i≤n h(x i), which implies that
Moreover, \({\boldsymbol \varLambda }_{nj}=Var(\sqrt {n}S_{n}^{j}({\boldsymbol \beta }_{0}))=n^{-1}\sum _{i=1}^{n}\eta _{ij}({\boldsymbol \beta }_{0})\eta _{ij}^{\tau }({\boldsymbol \beta }_{0})=\varSigma _{{\boldsymbol \beta }_{0}}^{j}+o_{p}(1)\) by assumption (I 6), from which \(\varSigma _{{\boldsymbol \beta }_{0}}^{j}\) is assumed to be positive definite. Hence, following the proof of Lemma 3.1, we have \(\varSigma _{jn}\varSigma _{jn}^{\tau }-{\boldsymbol \varLambda }_{nj}\to {0}\; a.s.\) Since \(\sqrt {n}S_{n}^{j}({\boldsymbol \beta }_{0})\xrightarrow {\mathcal {D}}N(0,\varSigma _{{\boldsymbol \beta }_{0}}^{j})\), we have \(\|S_{n}^{j}({\boldsymbol \beta }_{0})\|=O_{p}(n^{-1/2})\). Now from Eq. (3.4), using similar arguments as those in Owen (1990), it is obtained that ∥ξ∥ = O p(n −1∕2). On the other hand, performing a Taylor expansion to the right-hand side of Eq. (3.5) results in
where \(\displaystyle {{\boldsymbol \gamma }_{n}=O_{P}(1)\sum _{i=1}^{n}|{\boldsymbol \xi }^{\tau }\eta _{ij}({\boldsymbol \beta }_{0})|{ }^{3}}\). Now, using similar arguments as in Owen (2001), we have
Using Slutsky’s lemma, we have \(\displaystyle {\sqrt {n}{\boldsymbol \varLambda }_{nj}^{-1/2}S_{n}^{j}({\boldsymbol \beta }_{0})\xrightarrow {\mathcal {D}} N_{p}(0, I_{p})}\) as n →∞, and therefore,
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Bindele, H.F., Zhao, Y. (2018). Rank-Based Empirical Likelihood for Regression Models with Responses Missing at Random. In: Zhao, Y., Chen, DG. (eds) New Frontiers of Biostatistics and Bioinformatics. ICSA Book Series in Statistics. Springer, Cham. https://doi.org/10.1007/978-3-319-99389-8_3
Download citation
DOI: https://doi.org/10.1007/978-3-319-99389-8_3
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-99388-1
Online ISBN: 978-3-319-99389-8
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)