Browsing by Author "Lee, JC"

Now showing 1 - 20 of 40

A note on Bayesian estimation and prediction for the beta-binomial model
(1999)
Lee, JC
;
Lio, YL
;
統計學研究所
;
Institute of Statistics
The beta-binomial model which is generated by a simple mixture model has been widely applied in the social, physical, and health sciences. Lee and Sabavala (1987) proposed a Bayesian approach with a conjugate-type beta family of priors for suitably transformed parameters in the beta-binomial, and demonstrated the simulations for a special case of two trials. The main purpose of this paper is to extend the study of Lee and Sabavala (1987) by a numerical double integration. This method can be used for the case of general trials. When the number of trials is two, the results are similar to those from Lee and Sabavala (1987). However, the predictions for the real problems are much better than the results in Lee and Sabavala (1987).
A robust approach to t linear mixed models applied to multiple sclerosis data
(2006-04-30)
Lin, TI
;
Lee, JC
;
統計學研究所
;
資訊管理與財務金融系註：原資管所+財金所
;
Institute of Statistics
We discuss a robust extension of linear mixed models based on the multivariate t distribution. Since longitudinal data are successively collected over time and typically tend to be autocorrelated, we employ a parsimonious first-order autoregressive dependence structure for the within-subject errors. A score test statistic for testing the existence of autocorrelation among the within-subject errors is derived. Moreover, we develop an explicit scoring procedure for the maximum likelihood estimation with standard errors as a by-product. The technique for predicting future responses of a subject given past measurements is also investigated. Results are illustrated with real data from a multiple sclerosis clinical trial. Copyright (c) 2005 John Wiley & Sons, Ltd.
Active pulse structural control using artificial neural networks
(2000-08-01)
Hung, SL
;
Kao, CY
;
Lee, JC
;
土木工程學系
;
Department of Civil Engineering
Increasing interest has focused on applying active control systems to civil engineering structures subjected to dynamic loading. This study presents an active pulse control algorithm, termed the adaptive neural structural active pulse (ANSAP) controller, to control civil engineering structures under dynamic loading. The ANSAP controller minimizes structural cumulative responses during earthquakes by applying active pulse control forces. The effect of pulses is assumed to be delayed until just before the next sampling time so that the control force can be calculated in time and applied; the newly developed control strategy circumvents the effect of time delay due to the computation time. The ANSAP controller also circumvents the difficulty of obtaining system parameters of a real structure for the algorithm for active structural control. Illustrative examples reveal significant reductions in cumulative structural responses, which demonstrates the feasibility of using the adaptive artificial network for controlling civil engineering structures under dynamic loading.
Analysis of drug dissolution data
(1999-04-15)
Lee, JC
;
Chen, DT
;
Hung, HN
;
Chen, JJ
;
統計學研究所
Drug absorption in the human body depends on the dissolution rate of the drug. Suitable dissolution characteristics are important to ensure that the drug will achieve the desired therapeutic effects. To assess the similarity of dissolution rates of several drug lots, we apply a general growth curve model with different covariance structures. The Box-Cox power transformation and the naive log transformation are applied to a function of the dissolution rate. The predictive sample-reuse, or cross-validation, method is employed in selecting an appropriate model with best predictive accuracy. A testing procedure for examining the similarity among the drug lots is also conducted. A partially Bayesian approach is used for the assessment of dissolution equivalence. Copyright (C) 1999 John Wiley & Sons, Ltd.
Anatomy of zero-norm states in string theory
(2005-04-01)
Chan, CT
;
Lee, JC
;
Yi-Yang
;
電子物理學系
;
Department of Electrophysics
We calculate and identify the counterparts of zero-norm states in the old covariant first quantized (OCFQ) spectrum of open bosonic string in two other quantization schemes of string theory, namely, the light-cone Del Giudice-Di Vecchia-Fubine zero-norm states and the off-shell Becchi-Rouet-Stora-Tyutin (BRST) zero-norm states (with ghost) in the Witten string field theory (WSFT). In particular, special attention is paid to the interparticle zero-norm states in all quantization schemes. For the case of the off-shell BRST zero-norm states, we impose the no-ghost conditions and recover exactly two types of on-shell zero-norm states in the OCFQ string spectrum for the first few low-lying mass levels. We then show that off-shell gauge transformations of WSFT are identical to the on-shell stringy gauge symmetries generated by two types of zero-norm states in the generalized massive sigma-model approach of string theory. The high-energy limit of these stringy gauge symmetries was recently used to calculate the proportionality constants, conjectured by Gross, among high-energy scattering amplitudes of different string states. Based on these zero-norm state calculations, we have thus related gauge symmetry of WSFT to the high-energy stringy symmetry of Gross.
Bayesian analysis of a general growth curve model with predictions using power transformations and AR(1) autoregressive dependence
(2000-03-01)
Lee, JC
;
Liu, KC
;
統計學研究所
;
Institute of Statistics
In this paper, we consider a Bayesian analysis of the unbalanced (general) growth curve model with AR(1) autoregressive dependence, while applying the Box-Cox power transformations. We propose exact, simple and Markov chain Monte Carlo approximate parameter estimation and prediction of future values. Numerical results are illustrated with real and simulated data.
Bayesian analysis of a growth curve model with a general autoregressive covariance structure
(2000-12-01)
Lee, JC
;
Chang, CH
;
交大名義發表
;
National Chiao Tung University
In this paper wt: consider from maximum likelihood and Bayesian points of view the generalized growth curve model when the covariance matrix has a Toeplitz structure. This covariance is a generalization of the AR(I) dependence structure. Inferences on the parameters as well as the future values are Included. The results are illustrated with several real data sets.
Bayesian analysis of a growth curve model with power transformation, random effects and AR(1) dependence
(2001-02-01)
Lee, JC
;
Lien, WH
;
統計學研究所
;
Institute of Statistics
In this paper we devote ourselves to a general growth curve model with power transformation, random effects and AR(1) dependence via a Bayesian approach. Two priors are proposed and both parameter estimation and prediction of future values are considered. Some numerical results with a set of real data are also given.
Bayesian analysis of Box-Cox transformed linear mixed models with ARMA(p, q) dependence
(2005-08-01)
Lee, JC
;
Lin, TI
;
Lee, KJ
;
Hsu, YL
;
統計學研究所
In this paper, we present a Bayesian inference methodology for Box-Cox transformed linear mixed model with ARMA(p, q) errors using approximate Bayesian and Markov chain Monte Carlo methods. Two priors are proposed and put into comparisons in parameter estimation and prediction of future values. The advantages of Bayesian approach over maximum likelihood method are demonstrated by both real and simulated data. (c) 2004 Elsevier B.V. All rights reserved.
Bayesian analysis of growth curves with AR(1) dependence
(1997-11-15)
Lee, JC
;
Hsu, YL
;
統計學研究所
;
Institute of Statistics
In this paper we consider Bayesian analysis of the generalized growth curve model when the covariance matrix Sigma = sigma(2)C where C=(rho(i-j)), sigma(2) > 0 and -1 < rho < 1 are unknown. We consider both parameter estimation and prediction of future values. Results are illustrated with real and simulated data. (C) 1997 Elsevier Science B.V.
Bayesian estimation for time-series regressions improved with exact likelihoods
(2004-10-01)
Chen, CWS
;
Lee, JC
;
Lee, HY
;
Niu, WF
;
統計學研究所
We propose an estimation procedure for time-series regression models under the Bayesian inference framework. With the exact method of Wise [Wise, J. (1955). The autocorrelation function and spectral density function. Biometrika, 42, 151-159], an exact likelihood function can be obtained instead of the likelihood conditional on initial observations. The constraints on the parameter space arising from the stationarity conditions are handled by a reparametrization, which was not taken into consideration by Chib [Chib, S. (1993). Bayes regression with autoregressive errors: A Gibbs sampling approach. J. Econometrics, 58, 275-294] or Chib and Greenberg [Chib, S. and Greenberg, E. (1994). Bayes inference in regression model with ARMA(p, q) errors. J. Econometrics, 64, 183-206]. Simulation studies show that our method leads to better inferential results than their results.
Bayesian prediction analysis for growth curve model using noninformative priors
(2002-06-01)
Shieh, G
;
Lee, JC
;
統計學研究所
;
管理科學系
;
Institute of Statistics
We apply a Bayesian approach to the problem of prediction in an unbalanced growth curve model using noninformative priors. Due to the complexity of the model, no analytic forms of the predictive densities are available. We propose both approximations and a prediction-oriented Metropolis-Hastings sampling algorithm for two types of prediction, namely the prediction of future observations for a new subject and the prediction of future values for a partially observed subject. They are illustrated and compared through real data and simulation studies. Two of the approximations compare favorably with the approximation in Fearn (1975, Biometrika, 62, 89-100) and are very comparable to the more accurate Rao-Blackwellization from Metropolis-Hastings sampling algorithm.
Calculation of zero-norm states and reduction of stringy scattering amplitudes
(2005-07-01)
Lee, JC
;
電子物理學系
;
Department of Electrophysics
We give a simplified method to generate two types of zero-norm states in the old covariant first quantized (OCFQ) spectrum of open bosonic string. Zero-norm states up to the fourth massive level and general formulas of some zero-norm tensor states at arbitrary mass levels are calculated. On-shell Ward identities generated by zero-norm states and the factorization property of stringy vertex operators can then be used to argue that the string-tree scattering amplitudes of the degenerate lower spin propagating states are fixed by those of higher spin propagating states at each fixed mass level. This decoupling phenomenon is, in contrast to Gross's high-energy symmetries, valid to all energies. As examples, we explicitly demonstrate this stringy phenomenon up to the fourth massive level (spin-five), which justifies the calculation of two other previous approaches based on the massive worldsheet sigma-model and Witten's string field theory (WSFT).
Chan-Paton soliton gauge states of the compactified open string
(2000-04-01)
Lee, JC
;
電子物理學系
;
Department of Electrophysics
Mle study the mechanism of the enhanced gauge symmetry of the bosonic open string compactified on a torus by analyzing the zero-norm soliton (non-zero winding of the Wilson line) gauge states in the spectrum. Unlike the closed string case, we find that the soliton gauge state exists only at massive levels. These soliton gauge states correspond to the existence of enhanced massive gauge symmetries with transformation parameters containing both Einstein and Yang-Mills indices. In the T-dual picture, these symmetries exist only at some discrete values of compactified radii when N D-branes are coincident.
Consistent quantization of noncommutative D-brane in non-constant B field background
(2002-04-30)
Lee, JC
;
電子物理學系
;
Department of Electrophysics
We perform the consistent quantization of open string D-brane in non-constant NS-NS closed string B field background by directly imposing the worldsheet conformal symmetry. In addition to the previous noncommutative D-brane coordinates quantization with constant noncommutative paxameter theta, we obtain a set of constraints on the B field which can be interpreted as a quantum consistent deformation of a curved noncommutative D-brane.
Decoupling of degenerate positive-norm states in Witten's string field theory
(2003-04-15)
Kao, HC
;
Lee, JC
;
電子物理學系
;
Department of Electrophysics
We show that the degenerate positive-norm physical propagating fields of the open bosonic string can be gauged to the higher rank fields at the same mass level. As a result, their scattering amplitudes can be determined from those of the higher spin fields. This phenomenon arises from the existence of two types of zero-norm states with the same Young representations as those of the degenerate positive-norm states in the old covariant first quantized (OCFQ) spectrum. This is demonstrated by using the lowest order gauge transformation of Witten's string field theory (WSFT) up to the fourth massive level (spin-five), and is found to be consistent with conformal field theory calculation based on the first quantized generalized sigma-model approach. In particular, on-shell conditions of zero-norm states in the OCFQ stringy gauge transformation are found to correspond, in a one-to-one manner, to the background ghost fields in off-shell gauge transformation of WSFT. The implication of decoupling of scalar modes on Sen's conjectures is also briefly discussed.
Estimation and prediction of generalized growth curve with grouping variances in AR(q) dependence structure
(2003)
Lee, JC
;
Hsu, YL
;
統計學研究所
;
Institute of Statistics
In this paper we consider maximum likelihood analysis of generalized growth curve model with the Box-Cox transformation when the covariance matrix has AR(q) dependence structure with grouping variances. The covariance matrix under consideration is Sigma = DsigmaCDsigma where C is the correlation matrix with stationary autoregression process of order q, q < p and D-σ is a diagonal matrix with p elements divided into g(&LE; p) groups, i.e., D-σ is a function of {σ(1),..., σ(g)} and -1 < p < 1 and σ(l), l = 1,...g, are unknown. We consider both parameter estimation and prediction of future values. Results are illustrated with real and simulated data.
Exact tests in simple growth curve models and one-way ANOVA with equicorrelation error structure
(2003-02-01)
Lin, SH
;
Lee, JC
;
統計學研究所
;
Institute of Statistics
We consider exact tests with several equicorrelation error structures and combination of equicorrelation covariance structures in simple growth curve model having single or multiple treatments and in one-way ANOVA model. Exact inferences using generalized p-values are obtained. Tests for equal treatment effects under equal equicorrelation error term and for unequal equicorrelation error terms are also developed. Two examples are given to illustrate the importance of our results. According to our findings, we would be better off dropping the assumption of equal variance when the heteroscedasticity is serious. Therefore, tests based on generalized p-values without the assumption of equal variance are much more powerful than tests with this assumption. (C) 2003 Elsevier Science (USA). All rights reserved.
Fermonic zero-norm states and enlarged supersymmetries of Type II string
(2000-04-01)
Lee, JC
;
電子物理學系
;
Department of Electrophysics
We calculate the NS-R fermionic zero-norm states of the type II string spectrum. The massless and some possible massive zero-norm states are seen to be responsible for the space-time supersymmetry. The existence of other fermionic massive zero-norm states with higher spinor-tensor indices correspond to new enlarged boson-fermion symmetries of the theory at high energy. We also discuss the R-R charges and R-R zero-norm states and justify the idea that the perturbative string does not carry the massless R-R charges.
Generalized confidence intervals for the ratio of means of two normal populations
(2004-06-01)
Lee, JC
;
Lin, SH
;
統計學研究所
;
資訊管理與財務金融系註：原資管所+財金所
;
Institute of Statistics
Based on the generalized p-values and generalized confidence interval developed by Tsui and Weerahandi (J. Amer. Statist. Assoc. 84 (1989) 602), Weerahandi (J. Amer. Statist. Assoc. 88 (1993) 899), respectively, hypothesis testing and confidence intervals for the ratio of means of two normal populations are developed to solve Fieller's problems. We use two different procedures to find two potential generalized pivotal quantities. One procedure is to find the generalized pivotal quantity based directly on the ratio of means. The other is to treat the problem as a pseudo Behrens-Fisher problem through testing the two-sided hypothesis on 0, and then to construct the 1 - alpha generalized confidence interval as a counterpart of generalized p-values. Illustrative examples show that the two proposed methods arc numerically equivalent for large sample sizes. Furthermore, our simulation study shows that confidence intervals based on generalized p-values without the assumption of identical variance are more efficient than two other methods, especially in the situation in which the heteroscedasticity of the two populations is serious. (C) 2003 Elsevier B.V. All rights reserved.