Methods | hmax | NoDv |
[15] | 0.126 | 16 |
[20] | 0.577 | 75 |
[21] | 0.675 | 45 |
[22] | 0.728 | 45 |
[23] | 0.752 | 84 |
Theorem 3.1 | 1.141 | 71 |
Theoretical maximal value | 1.463 | − |
In this paper, we consider the multiple robust estimation of the parameters in the varying-coefficient partially linear model with response missing at random. The multiple robust estimation method is proposed, and the multiple robustness of the proposed method is proved. Numerical simulations are conducted to investigate the finite sample performance of the proposed estimators compared with other competitors.
Citation: Yaxin Zhao, Xiuli Wang. Multiple robust estimation of parameters in varying-coefficient partially linear model with response missing at random[J]. Mathematical Modelling and Control, 2022, 2(1): 24-33. doi: 10.3934/mmc.2022004
[1] | Patarawadee Prasertsang, Thongchai Botmart . Improvement of finite-time stability for delayed neural networks via a new Lyapunov-Krasovskii functional. AIMS Mathematics, 2021, 6(1): 998-1023. doi: 10.3934/math.2021060 |
[2] | Jenjira Thipcha, Presarin Tangsiridamrong, Thongchai Botmart, Boonyachat Meesuptong, M. Syed Ali, Pantiwa Srisilp, Kanit Mukdasai . Robust stability and passivity analysis for discrete-time neural networks with mixed time-varying delays via a new summation inequality. AIMS Mathematics, 2023, 8(2): 4973-5006. doi: 10.3934/math.2023249 |
[3] | Boonyachat Meesuptong, Peerapongpat Singkibud, Pantiwa Srisilp, Kanit Mukdasai . New delay-range-dependent exponential stability criterion and H∞ performance for neutral-type nonlinear system with mixed time-varying delays. AIMS Mathematics, 2023, 8(1): 691-712. doi: 10.3934/math.2023033 |
[4] | Yonggwon Lee, Yeongjae Kim, Seunghoon Lee, Junmin Park, Ohmin Kwon . An improved reachable set estimation for time-delay linear systems with peak-bounded inputs and polytopic uncertainties via augmented zero equality approach. AIMS Mathematics, 2023, 8(3): 5816-5837. doi: 10.3934/math.2023293 |
[5] | Rupak Datta, Ramasamy Saravanakumar, Rajeeb Dey, Baby Bhattacharya . Further results on stability analysis of Takagi–Sugeno fuzzy time-delay systems via improved Lyapunov–Krasovskii functional. AIMS Mathematics, 2022, 7(9): 16464-16481. doi: 10.3934/math.2022901 |
[6] | Yude Ji, Xitong Ma, Luyao Wang, Yanqing Xing . Novel stability criterion for linear system with two additive time-varying delays using general integral inequalities. AIMS Mathematics, 2021, 6(8): 8667-8680. doi: 10.3934/math.2021504 |
[7] | Xingyue Liu, Kaibo Shi . Further results on stability analysis of time-varying delay systems via novel integral inequalities and improved Lyapunov-Krasovskii functionals. AIMS Mathematics, 2022, 7(2): 1873-1895. doi: 10.3934/math.2022108 |
[8] | Xiao Ge, Xinzuo Ma, Yuanyuan Zhang, Han Xue, Seakweng Vong . Stability analysis of systems with additive time-varying delays via new bivariate quadratic reciprocally convex inequality. AIMS Mathematics, 2024, 9(12): 36273-36292. doi: 10.3934/math.20241721 |
[9] | Huahai Qiu, Li Wan, Zhigang Zhou, Qunjiao Zhang, Qinghua Zhou . Global exponential periodicity of nonlinear neural networks with multiple time-varying delays. AIMS Mathematics, 2023, 8(5): 12472-12485. doi: 10.3934/math.2023626 |
[10] | Wentao Le, Yucai Ding, Wenqing Wu, Hui Liu . New stability criteria for semi-Markov jump linear systems with time-varying delays. AIMS Mathematics, 2021, 6(5): 4447-4462. doi: 10.3934/math.2021263 |
In this paper, we consider the multiple robust estimation of the parameters in the varying-coefficient partially linear model with response missing at random. The multiple robust estimation method is proposed, and the multiple robustness of the proposed method is proved. Numerical simulations are conducted to investigate the finite sample performance of the proposed estimators compared with other competitors.
Over the last two decades, many researches used LKF method to get stability results for time-delay systems [1,2]. The LKF method has two important technical steps to reduce the conservatism of the stability conditions. The one is how to construct an appropriate LKF, and the other is how to estimate the derivative of the given LKF. For the first one, several types of LKF are introduced, such as integral delay partitioning method based on LKF [3], the simple LKF [4,5], delay partitioning based LKF [6], polynomial-type LKF [7], the augmented LKF [8,9,10]. The augmented LKF provides more freedom than the simple LKF in the stability criteria because of introducing several extra matrices. The delay partitioning based LKF method can obtain less conservative results due to introduce several extra matrices and state vectors. For the second step, several integral inequalities have been widely used, such as Jensen inequality [11,12,13,14], Wirtinger inequality [15,16], free-matrix-based integral inequality [17], Bessel-Legendre inequalities [18] and the further improvement of Jensen inequality [19,20,21,22,23,24,25]. The further improvement of Jensen inequality [22] is less conservative than other inequalities. However, The interaction between the delay partitioning method and the further improvement of Jensen inequality [23] was not considered fully, which may increase conservatism. Thus, there exists room for further improvement.
This paper further researches the stability of distributed time-delay systems and aims to obtain upper bounds of time-delay. A new LKF is introduced via the delay partitioning method. Then, a less conservative stability criterion is obtained by using the further improvement of Jensen inequality [22]. Finally, an example is provided to show the advantage of our stability criterion. The contributions of our paper are as follows:
∙ The integral inequality in [23] is more general than previous integral inequality. For r=0,1,2,3, the integral inequality in [23] includes those in [12,15,21,22] as special cases, respectively.
∙ An augmented LKF which contains general multiple integral terms is introduced to reduce the conservatism via a generalized delay partitioning approach. For example, the ∫tt−1mhx(s)ds, ∫tt−1mh∫tu1x(s)dsdu1, ⋯, ∫tt−1mh∫tu1⋯∫tuN−1x(s)dsduN−1⋯du1 are added as state vectors in the LKF, which may reduce the conservatism.
∙ In this paper, a new LKF is introduced based on the delay interval [0,h] is divided into m segments equally. From the LKF, we can conclude that the relationship among x(s), x(s−1mh) and x(s−m−1mh) is considered fully, which may yield less conservative results.
Notation: Throughout this paper, Rm denotes m-dimensional Euclidean space, A⊤ denotes the transpose of the A matrix, 0 denotes a zero matrix with appropriate dimensions.
Consider the following time-delay system:
˙x(t)=Ax(t)+B1x(t−h)+B2∫tt−hx(s)ds, | (2.1) |
x(t)=Φ(t),t∈[−h,0], | (2.2) |
where x(t)∈Rn is the state vector, A,B1,B2∈Rn×n are constant matrices. h>0 is a constant time-delay and Φ(t) is initial condition.
Lemma2.1. [23] For any matrix R>0 and a differentiable function x(s),s∈[a,b], the following inequality holds:
∫ba˙xT(s)R˙x(s)ds≥r∑n=0ρnb−aΦn(a,b)TRΦn(a,b), | (2.3) |
where
ρn=(n∑k=0cn,kn+k+1)−1, |
cn,k={1,k=n,n≥0,−n−1∑t=kf(n,t)ct,k,k=0,1,⋯n−1,n≥1, |
Φl(a,b)={x(b)−x(a),l=0,l∑k=0cl,kx(b)−cl,0x(a)−l∑k=1cl,kk!(b−a)kφk(a,b)x(t),l≥1, |
f(l,t)=t∑j=0ct,jl+j+1/t∑j=0ct,jt+j+1, |
φk(a,b)x(t)={∫bax(s)ds,k=1,∫ba∫bs1⋯∫bsk−1x(sk)dsk⋯ds2dss1,k>1. |
Remark2.1. The integral inequality in Lemma 2.1 is more general than previous integral inequality. For r=0,1,2,3, the integral inequality (2.3) includes those in [12,15,21,22] as special cases, respectively.
Theorem3.1. For given integers m>0,N>0, scalar h>0, system (2.1) is asymptotically stable, if there exist matrices P>0, Q>0, Ri>0,i=1,2,⋯,m, such that
Ψ=ξT1Pξ2+ξT2Pξ1+ξT3Qξ3−ξT4Qξ4+m∑i=1(hm)2ATdRiAd−m∑i=1r∑n=0ρnωn(t−imh,t−i−1mh)Ri×ωn(t−imh,t−i−1mh)<0, | (3.1) |
where
ξ1=[eT1ˉET0ˉET1ˉET2⋯ˉETN]T, |
ξ2=[ATdET0ET1ET2⋯ETN]T, |
ξ3=[eT1eT2⋯eTm]T, |
ξ4=[eT2eT3⋯eTm+1]T, |
ˉE0=hm[eT2eT3⋯eTm+1]T, |
ˉEi=hm[eTim+2eTim+3⋯eTim+m+1]T,i=1,2,⋯,N, |
Ei=hm[eT1−eTim+2eT2−eTim+3⋯eTm−eTm(i+1)+1]T,i=0,1,2,⋯,N, |
Ad=Ae1+B1em+1+B2m∑i=0em+1+i, |
ωn(t−imh,t−i−1mh)={ei−ei+i,n=0,n∑k=0cn,kei−cn,0ei+1−n∑k=1cn,kk!e(k−1)m+k+1,n≥1, |
ei=[0n×(i−1)nIn×n0n×(Nm+1−i)]T,i=1,2,⋯,Nm+1. |
Proof. Let an integer m>0, [0,h] can be decomposed into m segments equally, i.e., [0,h]=⋃mi=1[i−1mh,imh]. The system (2.1) is transformed into
˙x(t)=Ax(t)+B1x(t−h)+B2m∑i=1∫t−i−1mht−imhx(s)ds. | (3.2) |
Then, a new LKF is introduced as follows:
V(xt)=ηT(t)Pη(t)+∫tt−hmγT(s)Qγ(s)ds+m∑i=1hm∫−i−1mh−imh∫tt+v˙xT(s)Ri˙x(s)dsdv, | (3.3) |
where
η(t)=[xT(t)γT1(t)γT2(t)⋯γTN(t)]T, |
γ1(t)=[∫tt−1mhx(s)ds∫t−1mht−2mhx(s)ds⋮∫t−m−1mht−hx(s)ds],γ2(t)=mh[∫tt−1mh∫tu1x(s)dsdu1∫t−1mht−2mh∫t−1mhu1x(s)dsdu1⋮∫t−m−1mht−h∫t−m−1mhu1x(s)dsdu1],⋯, |
γN(t)=(mh)N−1×[∫tt−1mh∫tu1⋯∫tuN−1x(s)dsduN−1⋯du1∫t−1mht−2mh∫t−1mhu1⋯∫t−1mhuN−1x(s)dsduN−1⋯du1⋮∫t−m−1mht−h∫t−m−1mhu1⋯∫t−m−1mhuN−1x(s)dsduN−1⋯du1], |
γ(s)=[x(s)x(s−1mh)⋮x(s−m−1mh)]. |
The derivative of V(xt) is given by
˙V(xt)=2ηT(t)P˙η(t)+γT(t)Qγ(t)−γT(t−hm)Qx(t−hm)+m∑i=1(hm)2˙xT(t)Ri˙x(t)−m∑i=1hm∫t−i−1mht−imh˙xT(s)Ri˙x(s)ds. |
Then, one can obtain
˙V(xt)=ϕT(t){ξT1Pξ2+ξT2Pξ1+ξT3Qξ3−ξT4Qξ4+m∑i=1(hm)2ATdRiAd}ϕ(t)−m∑i=1hm∫t−i−1mht−imh˙xT(s)Ri˙x(s)ds, | (3.4) |
ϕ(t)=[xT(t)γT0(t)γT1(t)⋯γTN(t)]T, |
γ0(t)=[xT(t−1mh)xT(t−2mh)⋯xT(t−h)]T. |
By Lemma 2.1, one can obtain
−hm∫t−i−1mht−imh˙xT(s)Ri˙x(s)ds≤−r∑l=0ρlωl(t−imh,t−i−1mh)Ri×ωl(t−imh,t−i−1mh). | (3.5) |
Thus, we have ˙V(xt)≤ϕT(t)Ψϕ(t) by (3.4) and (3.5). We complete the proof.
Remark3.1. An augmented LKF which contains general multiple integral terms is introduced to reduce the conservatism via a generalized delay partitioning approach. For example, the ∫tt−1mhx(s)ds, ∫tt−1mh∫tu1x(s)dsdu1, ⋯, ∫tt−1mh∫tu1⋯∫tuN−1x(s)dsduN−1⋯du1 are added as state vectors in the LKF, which may reduce the conservatism.
Remark3.2. For r=0,1,2,3, the integral inequality (3.5) includes those in [12,15,21,22] as special cases, respectively. This may yield less conservative results. It is worth noting that the number of variables in our result is less than that in [23].
Remark3.3. Let B2=0, the system (2.1) can reduces to system (1) with N=1 in [23]. For m=1, the LKF in this paper can reduces to LKF in [23]. So the LKF in our paper is more general than that in [23].
This section gives a numerical example to test merits of our criterion.
Example 4.1. Consider system (2.1) with m=2,N=3 and
A=[01−100−1],B1=[0.00.10.10.2],B2=[0000]. |
Table 1 lists upper bounds of h by our methods and other methods in [15,20,21,22,23]. Table 1 shows that our method is more effective than those in [15,20,21,22,23]. It is worth noting that the number of variables in our result is less than that in [23]. Furthermore, let h=1.141 and x(0)=[0.2,−0.2]T, the state responses of system (1) are given in Figure 1. Figure 1 shows the system (2.1) is stable.
Methods | hmax | NoDv |
[15] | 0.126 | 16 |
[20] | 0.577 | 75 |
[21] | 0.675 | 45 |
[22] | 0.728 | 45 |
[23] | 0.752 | 84 |
Theorem 3.1 | 1.141 | 71 |
Theoretical maximal value | 1.463 | − |
In this paper, a new LKF is introduced via the delay partitioning method. Then, a less conservative stability criterion is obtained by using the further improvement of Jensen inequality. Finally, an example is provided to show the advantage of our stability criterion.
This work was supported by Basic Research Program of Guizhou Province (Qian Ke He JiChu[2021]YiBan 005); New Academic Talents and Innovation Program of Guizhou Province (Qian Ke He Pingtai Rencai[2017]5727-19); Project of Youth Science and Technology Talents of Guizhou Province (Qian Jiao He KY Zi[2020]095).
The authors declare that there are no conflicts of interest.
[1] |
J. Fan, T. Huang, Profile Likelihood Inferences on Semiparametric Varying-Coefficient Partially Linear Models, Bernoulli, 11 (2005), 1031–1057. https://doi.org/10.3150/bj/1137421639 doi: 10.3150/bj/1137421639
![]() |
[2] |
J. You, Y. Zhou, Empirical likelihood for semiparametric varying-coefficient partially linear regression models, Stat. Probabil. Lett., 76 (2006), 412–422. https://doi.org/10.1016/j.spl.2005.08.029 doi: 10.1016/j.spl.2005.08.029
![]() |
[3] |
Z. Huang, R. Zhang, Empirical likelihood for nonparametric parts in semiparametric varying-coefficient partially linear models, Stat. Probabil. Lett., 79 (2009), 1798–1808. https://doi.org/10.1016/j.spl.2009.05.008 doi: 10.1016/j.spl.2009.05.008
![]() |
[4] | P. Zhao, Infrerence for semiparametric varying coefficient partially linear models, doctoral dissertation, Beijing: Beijing University of Technology, 2010. |
[5] |
J. Feng, R. Zhang, Y. Lu, Inference on varying-coefficient partially linear regression model, Acta Math. Appl. Sin-E, 31 (2015), 139–156. https://doi.org/10.1007/s10255-015-0457-5 doi: 10.1007/s10255-015-0457-5
![]() |
[6] | R. Little, D. Rubin, Statistical analysis with missing data, 1 Eds., New York: Wiley Press, 1986. |
[7] |
D. Horvitz, D. Thompson, A generalization of sampling without replacement from a finite universe, J. Am. Stat. Assoc., 47 (1952), 663–685. https://doi.org/10.1080/01621459.1952.10483446 doi: 10.1080/01621459.1952.10483446
![]() |
[8] | F. Yates, The analysis of replicated experiments when the field results are incomplete, Exp. Agr., 1 (1933), 129–142. |
[9] |
P. Cheng, Nonparametric estimation of mean functionals with data missing at random, J. Am. Stat. Assoc., 89 (1994), 81–87. https://doi.org/10.1080/01621459.1994.10476448 doi: 10.1080/01621459.1994.10476448
![]() |
[10] |
Q. Wang, J. Rao, Empirical likelihood for linear regression models under imputation for missing responses, Can. J. Stat., 29 (2001), 596–608. https://doi.org/10.2307/3316009 doi: 10.2307/3316009
![]() |
[11] |
Q. Wang, J. Rao, Empirical likelihood-based inference under imputation for missing response data, Ann. Stat., 30 (2002), 896–924. https://doi.org/10.1214/aos/1028674845 doi: 10.1214/aos/1028674845
![]() |
[12] |
Q. Wang, O. Linton, W. Hardle, Semi-parametric regression analysis with missing response at random, J. Am. Stat. Assoc., 99 (2004), 334–345. https://doi.org/10.1198/016214504000000449 doi: 10.1198/016214504000000449
![]() |
[13] |
J. Robins, A. Rotnitzky, L. Zhao, Estimation of Regression Coefficients When Some Regressors Are Not Always Observed, J. Am. Stat. Assoc., 89 (1994), 846–866. https://doi.org/10.1080/01621459.1994.10476818 doi: 10.1080/01621459.1994.10476818
![]() |
[14] |
J. Robins, A. Rotnitzky, Semiparametric efficiency in multivariate regression models with missing data, J. Am. Stat. Assoc., 90 (1995), 122–129. https://doi.org/10.1080/01621459.1995.10476494 doi: 10.1080/01621459.1995.10476494
![]() |
[15] |
D. Scharfstein, A. Rotnitzky, R. Robins, Adjusting for Nonignorable Drop-Out Using Semiparametric Nonresponse Models, J. Am. Stat. Assoc., 94 (1999), 1096–1120. https://doi.org/10.1080/01621459.1999.10473862 doi: 10.1080/01621459.1999.10473862
![]() |
[16] |
J. Kang, J. Schafer, Demystifying double robustness: A comparison of alternative strategies for estimating a population mean from incomplete data (with Discussion), Statistics, 22 (2007), 523–539. https://doi.org/10.1214/07-STS227 doi: 10.1214/07-STS227
![]() |
[17] |
J. Qin, J. Shao, B. Zhang, Efficient and doubly robust imputation for covariate-dependent missing responses, J. Am. Stat. Assoc., 103 (2008), 797–810. https://doi.org/10.1198/016214508000000238 doi: 10.1198/016214508000000238
![]() |
[18] |
W. Cao, A. Tsiatis, M. Davidian, Improving efficiency and robustness of the doubly robust estimator for a population mean with incomplete data, Biometrika, 96 (2009), 723–734. https://doi.org/10.1093/biomet/asp033 doi: 10.1093/biomet/asp033
![]() |
[19] |
P. Han, A note on improving the efficiency of inverse probability weighted estimator using the augmentation term, Stat. Probabil. Lett., 82 (2012), 2221–2228. https://doi.org/10.1016/j.spl.2012.08.005 doi: 10.1016/j.spl.2012.08.005
![]() |
[20] |
A. Rotnitzky, Q. Lei, M. Sued, J. Robins, Improved double-robust estimation in missing data and causal inference models, Biometrika, 99 (2012), 439–456. https://doi.org/10.1093/biomet/ass013 doi: 10.1093/biomet/ass013
![]() |
[21] |
P. Han, L. Wang, Estimation with missing data: Beyond double robustness, Biometrika, 100 (2013), 417–430. https://doi.org/10.1093/biomet/ass087 doi: 10.1093/biomet/ass087
![]() |
[22] |
P. Han, Multiply Robust Estimation in Regression Analysis With Missing Data, J. Am. Stat. Assoc., 109 (2014), 1159–1173. https://doi.org/10.1080/01621459.2014.880058 doi: 10.1080/01621459.2014.880058
![]() |
[23] |
Y. Sun, L. Wang, P. Han, Multiply robust estimation in nonparametric regression with missing data, J. Nonparametr. Stat., 32 (2020), 73–92. https://doi.org/10.1080/10485252.2019.1700254 doi: 10.1080/10485252.2019.1700254
![]() |
[24] |
P. Han, Calibration and multiple robustness when data are missing not at random, Stat. Sinica, 109 (2018), 1725–1740. https://doi.org/10.5705/ss.202015.0408 doi: 10.5705/ss.202015.0408
![]() |
[25] |
W. Li, S. Yang, P. Han, Robust estimation for moment condition models with data missing not at random, J. Stat. Plan. Infer., 207 (2020), 246–254. https://doi.org/10.1016/j.jspi.2020.01.001 doi: 10.1016/j.jspi.2020.01.001
![]() |
[26] |
H. White, Maximum likelihood estimation of misspecified models, Econometrica, 50 (1982), 1–25. https://doi.org/10.2307/1912526 doi: 10.2307/1912526
![]() |
1. | Yanyan Sun, Xiaoting Bo, Wenyong Duan, Qun Lu, Stability analysis of load frequency control for power systems with interval time-varying delays, 2023, 10, 2296-598X, 10.3389/fenrg.2022.1008860 | |
2. | Xiao Ge, Xinzuo Ma, Yuanyuan Zhang, Han Xue, Seakweng Vong, Stability analysis of systems with additive time-varying delays via new bivariate quadratic reciprocally convex inequality, 2024, 9, 2473-6988, 36273, 10.3934/math.20241721 |