Censored regression ("Tobit") models have been in common use, and their linear hypothesis testings have been widely studied. However, the critical values of these tests are usually related to quantities of a...Censored regression ("Tobit") models have been in common use, and their linear hypothesis testings have been widely studied. However, the critical values of these tests are usually related to quantities of an unknown error distribution and estimators of nuisance parameters. In this paper, we propose a randomly weighting test statistic and take its conditional distribution as an approximation to null distribution of the test statistic. It is shown that, under both the null and local alternative hypotheses, conditionally asymptotic distribution of the randomly weighting test statistic is the same as the null distribution of the test statistic. Therefore, the critical values of the test statistic can be obtained by randomly weighting method without estimating the nuisance parameters. At the same time, we also achieve the weak consistency and asymptotic normality of the randomly weighting least absolute deviation estimate in censored regression model. Simulation studies illustrate that the per-formance of our proposed resampling test method is better than that of central chi-square distribution under the null hypothesis.展开更多
A general technique to obtain simple analytic approximations for the first kind of modified Bessel functions. The general procedure is shown, and the parameter determination is explained through the applications to th...A general technique to obtain simple analytic approximations for the first kind of modified Bessel functions. The general procedure is shown, and the parameter determination is explained through the applications to this particular case I1/6(x)and I1/7(x). In this way, it shows how to apply the technique to any particular orderν, in order to obtain an approximation valid for any positive value of the variable x. In the present method power series and asymptotic expansion are simultaneously used. The technique is an extension of the multipoint quasirational approximation method, MPQA. The main idea is to look for a bridge function between the power and asymptotic expansion of the I1/6(x), and similar procedure for I1/7(x). To perform this, rational functions are combined with hyperbolic ones and fractional powers. The number of parameters to be determined for each case is four. The maximum relative errors are 0.0049 for ν=1/6, and 0.0047 for ν=7. However, these relative errors decrease outside of the small region of the variables, wherein the maximum relative errors are reached. There is a clear advantage of this procedure compared with any other ones.展开更多
The P-type update law has been the mainstream technique used in iterative learning control(ILC)systems,which resembles linear feedback control with asymptotical convergence.In recent years,finite-time control strategi...The P-type update law has been the mainstream technique used in iterative learning control(ILC)systems,which resembles linear feedback control with asymptotical convergence.In recent years,finite-time control strategies such as terminal sliding mode control have been shown to be effective in ramping up convergence speed by introducing fractional power with feedback.In this paper,we show that such mechanism can equally ramp up the learning speed in ILC systems.We first propose a fractional power update rule for ILC of single-input-single-output linear systems.A nonlinear error dynamics is constructed along the iteration axis to illustrate the evolutionary converging process.Using the nonlinear mapping approach,fast convergence towards the limit cycles of tracking errors inherently existing in ILC systems is proven.The limit cycles are shown to be tunable to determine the steady states.Numerical simulations are provided to verify the theoretical results.展开更多
Suppose that Y follows a χ^(p)-distribution with n degrees of freedom, and Z is the standardized form of Y. Let f(z,n,p) and F(z,n,p) denote the density function and the distribution function of Z, respectively. In t...Suppose that Y follows a χ^(p)-distribution with n degrees of freedom, and Z is the standardized form of Y. Let f(z,n,p) and F(z,n,p) denote the density function and the distribution function of Z, respectively. In this paper, we obtain the asymptotic expansion for f(z,n,p) and F(z,n,p). The validity of these results is illuminated by some numerical examples. We also investigate the power function of χ^(p)-test by the asymptotic expansion.展开更多
This paper presents an approach for estimating power of the score test, based on an asymptotic approximation to the power of the score test under contiguous alternatives. The method is applied to the problem of power ...This paper presents an approach for estimating power of the score test, based on an asymptotic approximation to the power of the score test under contiguous alternatives. The method is applied to the problem of power calculations for the score test of heteroscedasticity in European rabbit data （Ratkowsky, 1983）. Simulation studies are presented which indicate that the asymptotic approximation to the finite-sample situation is good over a wide range of parameter configurations.展开更多
In this paper we develop tests using L-statistic for constant relativefailure rate against monotone relative failure rate.Scale invariance,monotonepower and locally asymptotic power results are obtained.Some examples ...In this paper we develop tests using L-statistic for constant relativefailure rate against monotone relative failure rate.Scale invariance,monotonepower and locally asymptotic power results are obtained.Some examples arepresented.展开更多
考虑了乘积多项抽样下的对数线性模型.在这个模型下,文献[Jin Y H,Wu Y H.Mini mumφ-divergence esti mator and hierarchical testing in log-linear models under product-multinomial sampling.Journal of Statistical Planning a...考虑了乘积多项抽样下的对数线性模型.在这个模型下,文献[Jin Y H,Wu Y H.Mini mumφ-divergence esti mator and hierarchical testing in log-linear models under product-multinomial sampling.Journal of Statistical Planning and Inference,2009,139:3 488-3 500]用基于-散度和最小-散度估计构造的统计量研究了几类假设检验问题,这其中就有嵌套假设.最小-散度估计是极大似然估计的推广.在上述文献的基础上,给出了其中一类检验的功效函数的渐近逼近公式;另外,还研究了在一列近邻假设下检验统计量的渐近分布.通过模拟研究发现,与Pearson型统计量和对数极大似然比统计量相比,Cressie-Read型检验统计量有差不多的甚至更好的模拟功效和水平.展开更多
The higher order asymptotic fields at the tip of a sharp V-notchin a power-hardening material for plane strain problem of Mode I arederived. The order hierarchy in powers of r for various hardeningexponents n and notc...The higher order asymptotic fields at the tip of a sharp V-notchin a power-hardening material for plane strain problem of Mode I arederived. The order hierarchy in powers of r for various hardeningexponents n and notch angles β is obtained. The angulardistributions of stress for several cases are plotted. Theself-similarity behavior between the higher order terms is noticed.It is found that the terms with higher Order can be neglected for theV-notch angle β>45°.展开更多
A family of integral-type goodness-of-fit tests is investigated. This family includes some existing tests, such as the Cramer-von Mises test and Anderson-Darling test, etc. The asymptotic distributions of the tests in...A family of integral-type goodness-of-fit tests is investigated. This family includes some existing tests, such as the Cramer-von Mises test and Anderson-Darling test, etc. The asymptotic distributions of the tests in the family under the null and local alternative hypotheses are established. The almost sure convergence under a fixed underlying distribution is obtained. Furthermore, simulations are conducted to compare the powers of the tests in the family. Simulation results show that for different alternatives, the more powerful tests are different, and the parameter ）, has great influence on the tests in small sample cases.展开更多
A procedure is developed for power analysis and sample size calculation for a class of complex testing problems regarding the largest binomial probability under a combination of treatments.It is shown that the asympto...A procedure is developed for power analysis and sample size calculation for a class of complex testing problems regarding the largest binomial probability under a combination of treatments.It is shown that the asymptotic null distribution of the likelihood-ratio statistic is not parameterfree,but χ_(1)^(2) is a conservative asymptotic null distribution.A nonlinear Gauss-Seidel algorithm is proposed to uniquely determine the alternative for the power and sample size calculation given the baseline binomial probability.An example from an animal clinical trial is discussed.展开更多
The main purpose of this paper is to study the mean value properties of the character sums over the interval [1,p/8） by using the mean value theorems of the Dirichlet L-functions, and give an interesting mean value f...The main purpose of this paper is to study the mean value properties of the character sums over the interval [1,p/8） by using the mean value theorems of the Dirichlet L-functions, and give an interesting mean value formula for this study.展开更多
基金supported by National Natural Science Foundation of China (Grant No. 10471136)PhD Program Foundation of the Ministry of Education of ChinaSpecial Foundations of the Chinese Academy of Sciences and University of Science and Technology of China
文摘Censored regression ("Tobit") models have been in common use, and their linear hypothesis testings have been widely studied. However, the critical values of these tests are usually related to quantities of an unknown error distribution and estimators of nuisance parameters. In this paper, we propose a randomly weighting test statistic and take its conditional distribution as an approximation to null distribution of the test statistic. It is shown that, under both the null and local alternative hypotheses, conditionally asymptotic distribution of the randomly weighting test statistic is the same as the null distribution of the test statistic. Therefore, the critical values of the test statistic can be obtained by randomly weighting method without estimating the nuisance parameters. At the same time, we also achieve the weak consistency and asymptotic normality of the randomly weighting least absolute deviation estimate in censored regression model. Simulation studies illustrate that the per-formance of our proposed resampling test method is better than that of central chi-square distribution under the null hypothesis.
文摘A general technique to obtain simple analytic approximations for the first kind of modified Bessel functions. The general procedure is shown, and the parameter determination is explained through the applications to this particular case I1/6(x)and I1/7(x). In this way, it shows how to apply the technique to any particular orderν, in order to obtain an approximation valid for any positive value of the variable x. In the present method power series and asymptotic expansion are simultaneously used. The technique is an extension of the multipoint quasirational approximation method, MPQA. The main idea is to look for a bridge function between the power and asymptotic expansion of the I1/6(x), and similar procedure for I1/7(x). To perform this, rational functions are combined with hyperbolic ones and fractional powers. The number of parameters to be determined for each case is four. The maximum relative errors are 0.0049 for ν=1/6, and 0.0047 for ν=7. However, these relative errors decrease outside of the small region of the variables, wherein the maximum relative errors are reached. There is a clear advantage of this procedure compared with any other ones.
基金supported by the National Natural Science Foundation of China(62173333)Australian Research Council Discovery Program(DP200101199)。
文摘The P-type update law has been the mainstream technique used in iterative learning control(ILC)systems,which resembles linear feedback control with asymptotical convergence.In recent years,finite-time control strategies such as terminal sliding mode control have been shown to be effective in ramping up convergence speed by introducing fractional power with feedback.In this paper,we show that such mechanism can equally ramp up the learning speed in ILC systems.We first propose a fractional power update rule for ILC of single-input-single-output linear systems.A nonlinear error dynamics is constructed along the iteration axis to illustrate the evolutionary converging process.Using the nonlinear mapping approach,fast convergence towards the limit cycles of tracking errors inherently existing in ILC systems is proven.The limit cycles are shown to be tunable to determine the steady states.Numerical simulations are provided to verify the theoretical results.
基金Joint supported by Hubei Provincial Natural Science Foundation and Huangshi of China (2022CFD042)。
文摘Suppose that Y follows a χ^(p)-distribution with n degrees of freedom, and Z is the standardized form of Y. Let f(z,n,p) and F(z,n,p) denote the density function and the distribution function of Z, respectively. In this paper, we obtain the asymptotic expansion for f(z,n,p) and F(z,n,p). The validity of these results is illuminated by some numerical examples. We also investigate the power function of χ^(p)-test by the asymptotic expansion.
基金Supported by SSFC（04BTJ002）,the National Natural Science Foundation of China（10371016） and the Post-Doctorial Grant in Southeast University.
文摘This paper presents an approach for estimating power of the score test, based on an asymptotic approximation to the power of the score test under contiguous alternatives. The method is applied to the problem of power calculations for the score test of heteroscedasticity in European rabbit data （Ratkowsky, 1983）. Simulation studies are presented which indicate that the asymptotic approximation to the finite-sample situation is good over a wide range of parameter configurations.
基金The project was supported in part by the government of China,the Office of Naval Research grant N000 14-80-C-0163
文摘In this paper we develop tests using L-statistic for constant relativefailure rate against monotone relative failure rate.Scale invariance,monotonepower and locally asymptotic power results are obtained.Some examples arepresented.
基金Supported by National Natural Science Foundation of China(10871188,10801123)the Knowledge Innovation Programof the Chinese Academy of Sciences(KJCX3-SYW-S02)
文摘考虑了乘积多项抽样下的对数线性模型.在这个模型下,文献[Jin Y H,Wu Y H.Mini mumφ-divergence esti mator and hierarchical testing in log-linear models under product-multinomial sampling.Journal of Statistical Planning and Inference,2009,139:3 488-3 500]用基于-散度和最小-散度估计构造的统计量研究了几类假设检验问题,这其中就有嵌套假设.最小-散度估计是极大似然估计的推广.在上述文献的基础上,给出了其中一类检验的功效函数的渐近逼近公式;另外,还研究了在一列近邻假设下检验统计量的渐近分布.通过模拟研究发现,与Pearson型统计量和对数极大似然比统计量相比,Cressie-Read型检验统计量有差不多的甚至更好的模拟功效和水平.
基金the National Natural Science Foundation of China (Nos.10132010 and 10072033).
文摘The higher order asymptotic fields at the tip of a sharp V-notchin a power-hardening material for plane strain problem of Mode I arederived. The order hierarchy in powers of r for various hardeningexponents n and notch angles β is obtained. The angulardistributions of stress for several cases are plotted. Theself-similarity behavior between the higher order terms is noticed.It is found that the terms with higher Order can be neglected for theV-notch angle β>45°.
基金supported by the National Natural Science Foundation of China under Grant Nos. 10661003, 10371126the Guangxi Science Foundation under Grant No. 0832102the Doctor Foundation of Guangxi Normal University
文摘A family of integral-type goodness-of-fit tests is investigated. This family includes some existing tests, such as the Cramer-von Mises test and Anderson-Darling test, etc. The asymptotic distributions of the tests in the family under the null and local alternative hypotheses are established. The almost sure convergence under a fixed underlying distribution is obtained. Furthermore, simulations are conducted to compare the powers of the tests in the family. Simulation results show that for different alternatives, the more powerful tests are different, and the parameter ）, has great influence on the tests in small sample cases.
基金partially supported by the National Institutes of Health(NIH)grant R01-GM085205A1partially supported by the National Science Foundation(NSF)grant SES-1118469partially supported by the National Science Foundation(NSF)grant SES-1121794.
文摘A procedure is developed for power analysis and sample size calculation for a class of complex testing problems regarding the largest binomial probability under a combination of treatments.It is shown that the asymptotic null distribution of the likelihood-ratio statistic is not parameterfree,but χ_(1)^(2) is a conservative asymptotic null distribution.A nonlinear Gauss-Seidel algorithm is proposed to uniquely determine the alternative for the power and sample size calculation given the baseline binomial probability.An example from an animal clinical trial is discussed.
文摘The main purpose of this paper is to study the mean value properties of the character sums over the interval [1,p/8） by using the mean value theorems of the Dirichlet L-functions, and give an interesting mean value formula for this study.
