Abstract
We propose a kernel estimator of a hazard ratio that is based on a modification of Ćwik and Mielniczuk (Commun Stat-Theory Methods 18(8):3057–3069, 1989)’s method. A naive nonparametric estimator is Watson and Leadbetter (Sankhyā: Indian J Stat Ser A 26(1):101–116, 1964)’s one, which is naturally given by the kernel density estimator and the empirical distribution estimator. We compare the asymptotic mean squared error (AMSE) of the hazard estimators, and then, it is shown that the asymptotic variance of the new estimator is usually smaller than that of the naive one. We also discuss bias reduction of the proposed estimator and derived some modified estimators. While the modified estimators do not lose nonnegativity, their AMSE is small both theoretically and numerically.
Similar content being viewed by others
References
Chen, S. M., Hsu, Y. S., Liaw, J. T. (2009). On kernel estimators of density ratio. Statistics, 43(5), 463–479.
Ćwik, J., Mielniczuk, J. (1989). Estimating density ratio with application to discriminant analysis. Communications in Statistics-Theory and Methods, 18(8), 3057–3069.
De Haan, L., Ferreira, A. (2007). Extreme value theory: An introduction. New York, NY: Springer.
Funke, B., Kawka, R. (2015). Nonparametric density estimation for multivariate bounded data using two non-negative multiplicative bias correction methods. Computational Statistics & Data Analysis, 92, 148–162.
Gumbel, E. (1958). Statistics of extremes. New York, NY: Columbia University Press.
Hirukawa, M., Sakudo, M. (2014). Nonnegative bias reduction methods for density estimation using asymmetric kernels. Computational Statistics & Data Analysis, 75, 112–123.
Jones, M., Signorini, D. (1997). A comparison of higher-order bias kernel density estimators. Journal of the American Statistical Association, 92(439), 1063–1073.
Jones, M. C., Linton, O., Nielsen, J. P. (1995). A simple bias reduction method for density estimation. Biometrika, 82(2), 327.
McCune, E., McCune, S. (1987). On improving convergence rates for nonnegative kernel failure-rate function estimators. Statistics & Probability Letters, 6(2), 71–76.
Müller, H. G., Wang, J. L. (1994). Hazard rate estimation under random censoring with varying kernels and bandwidths. Biometrics, 50(1), 61.
Murthy, V. (1965). Estimation of jumps, reliability and hazard rate. The Annals of Mathematical Statistics, 36(3), 1032–1040.
Nielsen, J. P. (1998). Multiplicative bias correction in kernel hazard estimation. Scandinavian Journal of Statistics, 25(3), 541–553.
Patil, P. (1993). On the least squares cross-validation bandwidth in hazard rate estimation. The Annals of Statistics, 21(4), 1792–1810.
Quintela-del Río, A. (2007). Plug-in bandwidth selection in kernel hazard estimation from dependent data. Computational Statistics & Data Analysis, 51(12), 5800–5812.
Rosenblatt, M. (1956). Remarks on some nonparametric estimates of a density function. The Annals of Mathematical Statistics, 27(3), 832–837.
Ruppert, D., Cline, D. B. H. (1994). Bias reduction in kernel density estimation by smoothed empirical transformations. The Annals of Statistics, 22(1), 185–210.
Tanner, M. A., Wong, W. H. (1983). The estimation of the hazard function from randomly censored data by the kernel method. The Annals of Statistics, 11(3), 989–993.
Terrell, G. R., Scott, D. W. (1980). On improving convergence rates for nonnegative kernel density estimators. The Annals of Statistics, 8(5), 1160–1163.
Vieu, P. (1991a). Nonparametric regression: Optimal local bandwidth choice. Journal of the Royal Statistical Society Series B (Methodological), 53(2), 453–464.
Vieu, P. (1991b). Quadratic errors for nonparametric estimates under dependence. Journal of Multivariate Analysis, 39(2), 324–347.
Watson, G., Leadbetter, M. (1964). Hazard analysis ii. Sankhyā: The Indian Journal of Statistics, Series A26(1), 101–116.
Acknowledgements
The authors would like to appreciate the editor’s and referees’ valuable comments that helped us to improve this manuscript significantly. The authors gratefully acknowledge JSPS KAKENHI Grant Number JP15K11995 and JP16H02790.
Author information
Authors and Affiliations
Corresponding author
Appendices: Proofs of Theorems
Appendices: Proofs of Theorems
Proof of Theorem 1
For simplicity, we will use the following notation,
To begin with, we consider the following stochastic expansion of the direct estimator:
where
and \(M_n^*(w)\) is a r.v. between \(M_n(w)\) and M(w) with probability 1.
The main term of the expectation of \(\widehat{H}(x_0)\) is given by \(J_1\), as we will show the later. Since \(J_1\) is a sum of i.i.d. random variables, the expectation can be obtained directly:
Combining the following second moment,
we get the variance,
Next, we consider the following representation of \(J_2\)
where
Using the conditional expectation, we get the following equation:
Next, we have
After taking the conditional expectation, we find that if all of the (i, j, s, t) are different,
and
the term in which \(j=t\) and all of the (i, j, s) are different is the main term of \(E[J_2^2]\). If \(j=t\) and all of the (i, j, s) are different, we have
Using the conditional expectation of \(Q(X_i,X_j) Q(X_s,X_j)\) given \(X_i\) and \(X_s\), we find that
Therefore, the entire expectation of the last row is
Finally, we get
After similar calculations of the other terms, we find that if \(j=t\) and all of the (i, j, s) are different,
In addition, it is easy to see that the expectations of the other combinations of (i, j, s, t) are \(o(n^{-1})\). As a result, we have
The moments of \(J_3\) can be obtained in a similar manner; we find that
Finally, we will obtain upper bounds of \(|E[J_4^*]|\) and \(E[{(J_4^*)}^2]\). From the assumption of Theorem 1, we can see
Similarly, it follows that
To sum up, we conclude that \(J_2 + J_3 + J_4^*\) is asymptotically negligible for fixed \(x_0\). The main bias of \(\widehat{H}(x_0)\) comes from \(J_1\). From the Cauchy–Schwarz inequality, we find that the main term of the variance coincides with \(V[J_1]\). Now, we can get the AMSE of the direct estimator and prove Theorem 1. \(\square \)
Proof of Theorems 2 and 3
It follows from the above discussion that
Since \(J_1\) is a sum of i.i.d. random variables and the expectation of the second term is o(1), asymptotic normality holds for Theorem 2. \(\square \)
Furthermore, we can show that
and we can directly get Theorem 3 by taking a Taylor expansion. \(\square \)
Proof of Theorem 4
We follow the proof of Jones et al. (1995). \(\widehat{H}_{N}(x_0)\) is written as
It follows from the asymptotic expansion that
By taking the expectation of the ith term in this sum conditional on \(X_i\), we have
Thus, we have
It follows that
From the proof of Theorem 1, we can see that the following approximation holds:
\(H_N(x_0)\) can be seen as a Jones et al. (1995)’s density estimate for the random variable M(X) at the point \(M(x_0)\). Then, the asymptotic variance is given by
(see Jones et al. 1995). \(\square \)
Proof of Theorem 5
To obtain the asymptotic variance of the modified estimator \(\widehat{H}^{\dagger }\), we need to calculate the covariance term \(Cov[\widehat{H}_h(x_0), \widehat{H}_{2h}(x_0)]\) as shown in Sect. 4. From the proof of Theorem 1, it is easy to see
Consequently, we have
and the desired result. \(\square \)
About this article
Cite this article
Moriyama, T., Maesono, Y. New kernel estimators of the hazard ratio and their asymptotic properties. Ann Inst Stat Math 72, 187–211 (2020). https://doi.org/10.1007/s10463-018-0685-6
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-018-0685-6