臺灣博碩士論文加值系統

本論文針對聯合型II逐步設限策略下，討論韋伯(Weibull)、逆高斯(inverse Gaussian)、和Birnbaum-Saunders分配的參數估計，進而可以估計P(Y<X)。參數估計的求法主要根據三種方法:以牛頓(Newton-Raphson)法解概似方程式，以及隨機EM演算法和雙重退火演算法最佳化概似函數。因為韋伯(Weibull)分配的概似方程式只有一個未知變數，所以也可用二分(Bisection)法來解概似方程式。我們特別針對Weibull分配進行不同參數組合在四種方法下各種設限策略的參數估計模擬研究，並發現使用雙重退火算法所得的最大參數估計值的平均誤差與均方誤差(MSE)的表現是最佳。相同的結論也可在inverse Gaussian和Birnbaum-Saunders分配下得到。最後，我們用所提出的方法來估計一個真實數值例子之P(Y<X)。

In this thesis, the estimation of parameters of a certain family of two-parameter lifetime distributions based on joint progressively Type-II right-censored sample is studied. This family includes the Weibull, inverse Gaussian, and Birnbaum-Saunders distributions. Different numerical methods are used to compute the maximum likelihood estimates of the unknown parameters. These methods include the Newton-Raphson method, stochastic EM algorithm, and dual annealing algorithm with Python. These methods are compared via a Monte Carlo simulation study in terms of their biases and mean squared errors. Recommendations are made from the simulated results and a real data set has been analyzed for illustrative purposes.

1 Introduction 1
2 Likelihood Estimation of Model Parameters 6
2.1 Numerical Methods 7
2.1.1 Newton-Raphson Method 7
2.2 Stochastic Expectation-Maximization Algorithm 7
2.3 Dual Annealing Algorithm 9
2.4 Bisection Method Algorithm 11
3 Parameter Estimation for the Weibull Lifetime istributions 12
3.1 Likelihood Equations 12
3.2 The M-step in SEM Algorithm 14
4 Parameter Estimation for the Inverse Gaussian Lifetime Distributions 16
4.1 Likelihood Equations 16
4.2 The M-step in SEM Algorithm 17
5 Parameter Estimation for the Birnbaum-Saunders Lifetime Distributions 19
5.1 Likelihood Equations 19
5.2 The M-step in SEM Algorithm 20
6 Numerical Analysis 22
6.1 Simulation Study 22
6.2 An Illustrative Example 29
7 Concluding Remarks 32
8 Appendix 33
9 References 38

List of Tables
Table 1 Average values of biases and MSE of the MLE of the parameters in the Weibull distributions with μ1 = 0.5, λ1 = 0.6, μ2 = 0.6, λ2 = 0.7 . . . . . . . . . . . 23
Table 2 Average values of biases and MSE of the MLE of the parameters in the Weibull distributions with μ1 = 1.3, λ1 = 0.5, μ2 = 1.1, λ2 = 1.2 . . . . . . . . . . . 24
Table 3 Average values of biases and MSE of the MLE of the parameters in the Weibull distributions with μ = 0.5, λ1 = 0.5, λ2 = 1 . . . . . . . . . . . . . . . . . . . . . 25
Table 4 Average values of biases and MSE of the MLE of the parameters in the Weibull distributions with μ = 1, λ1 = 0.5, λ2 = 2 . . . . . . . . . . . . . . . . . . . . . . .26
Table 5 Average values of biases and MSE of the MLE of the parameters in the inverse Gaussian Lifetime distributions with μ1 = 0.5, λ1 = 1.5, μ2 = 0.6, λ2 = 1.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Table 6 Average values of biases and MSE of the MLE of the parameters in the Birnbaum-Saunders Lifetime distributions with μ1 = 1, λ1 = 0.8, μ2 = 0.5, λ2 = 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Table 7 Breakdown times of the insulating fluid at two voltage levels, and the corresponding MLE of μ and λ, log-likelihood, KS test statistics, and p values of the fitted Weibull models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Table 8 p values of the LR and KS tests for common shape parameter . . . . . 29
Table 9 The generated JPC samples based on four schemes . . . . . . . . . . . . 30
Table 10 The MLE of Q for different censoring schemes . . . . . . . . . . . . . . . . .31
Table 11 The 95% approximate, Boot-p, and Boot-t confidence intervals and widths for Q . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

List of Figures
Figure 1 Schematic representation of JPC scheme . . . . . . . . . . . . . . . . . . . . . 2

References

Abdel-Aty, Y. (2017). Exact likelihood inference for two populations from two-parameter exponential distributions under joint Type-II censoring. Communications in Statistics –Theory and Methods 46, 9026–9041.

Ahmad, K. E., Fakhry, M. E. and Jaheen, Z. F. (1997). Empirical Bayes estimation of P(Y < X) and characterizations of Burr-type X model. Journal of Statistical Planning and Inference 64, 297–308.

Ashour, S. K. and Abo-Kasem, O. E. (2014). Parameter estimation for two Weibull populations under joint Type II censored scheme. International Journal of Engineering 5, 31–36.

Balakrishnan, N. and Rasouli, A. (2008). Exact likelihood inference for two exponential populations under joint Type-II censoring. Computational Statistics & Data Analysis 52, 2725–2738.

Brownlee, J. (2021). Dual annealing optimization with Python. Machine Learning Mastery. https://machinelearningmastery.com/dual-annealing-optimization-with-python.

Celeux, G. and Diebolt, J. (1985). The SEM algorithm: a probabilistic teacher algorithm derived from the EM algorithm for the mixture problem. Computational Statistics Quarterly 2, 73–82.

Cariou, C. and Chehdi, K. (2008). Unsupervised texture segmentation/classification using 2-D autoregressive modeling and the stochastic expectation maximization algorithm. Pattern Recognition Letters 29, 905–917.

Constantine, K. and Karson, M. (1986). Estimation of P(Y < X) in the gamma case. Communications in Statistics – Simulation and Computation 15, 365–388.

Delignon, Y., Marzouki, A. and Pieczynski, W. (1997). Estimation of generalized mixtures and its application in image segmentation. IEEE Transactions on Image Processing 6, 1364–1375.

Diebolt, J. and Celeux, G. (1993). Asymptotic properties of a stochastic EM algorithm for estimating mixing proportions. Stochastic Models 9, 599– 613.

Diebolt, J. and Ip, E. H. S. (1996). Stochastic EM: Method and application. In Markov chain Monte Carlo in practice, pp. 259–273. London: Chapman & Hall/CRC.

Efron, B. and Tibshirani, R. J. (1994). An Introduction to the Bootstrap. New York: Chapman & Hall/CRC Press.

Kirkpatrick, S., Gelatt, Jr. C. D. and Vecchi, M.P. (1983). Optimization by simulated annealing. Science 220, 671–680.

Kundu, D. and Gupta, R. D. (2005). Estimation of P[Y < X] for generalized exponential distribution. Metrika 61, 291–308.

Kundu, D. and Gupta, R. D. (2006). Estimation of P[Y < X] for Weibull distributions. IEEE Transactions on Reliability 55, 270–280.

Lawless, J. F. (2003). Statistical Models and Methods for Lifetime Data. Second edition. New York: John Wiley.

Lin, C. T. and Ke, S. C. (2013). Estimation of P(Y < X) for location-scale distributions under joint progressively Type-II right censoring. Quality Technology & Quantitative Management, 10, 339–352.

Mondal, S. and Kundu, D. (2019a). Point and interval estimation of Weibull parameters based on joint progressively censored data. Sankhya B 81, 1–25.

Mondal, S. and Kundu, D. (2019b). A new two sample type-II progressive censoring scheme. Communications in Statistics – Theory and Methods, 48, 2602–2618.

Mondal, S. and Kundu, D. (2020) On the joint Type-II progressive censoring scheme. Communications in Statistics – Theory and Methods, 49, 958–976.

Nadar, M., Kizilaslan, F. and Papadopoulos, A. (2014). Classical and Bayesian estimation of P(Y < X) for Kumaraswamy’s distribution. Journal of Statistical Computation and Simulation, 84, 1505–1529.

Nelson, W. B. (1970). Statistical methods for accelerated lifetest data–the inverse power law model. General Electric Co. Technical Report 71-C-011, Schenectady, New York.

Nielsen, F. S. (2000). The stochastic EM algorithm: estimation and asymptotic results. Bernoulli 6, 457–489.

Parsi, S. and Bairamov, I. (2009). Expected values of the number of failures for two populations under joint Type-II progressive censoring. Computational Statistics & Data Analysis 53, 3560–3570.

Parsi, S., Ganjali, M. and Farsipour. N. S. (2011). Conditional maximum likelihood and interval estimation for two Weibull populations under joint Type-II progressive censoring. Communications in Statistics – Theory and Methods 40, 2117–2135.

Raqab, M. Z. and Kundu, D. (2005). Comparison of different estimators of P[Y < X] for a scaled Burr type X distribution. Communications in Statistics – Simulation and Computation 34, 465–483.

Rasouli, A. and Balakrishnan, N. (2010). Exact likelihood inference for two exponential populations under joint progressive Type-II censoring. Communications in Statistics – Theory and Methods 39, 2172–2191.

Shafay, A. R., Balakrishnan, N. and Aty, A. (2014). Bayesian inference based on a jointly type-II censored sample from two exponential populations. Journal of Statistical Computation and Simulation 84, 2427–2440.

Shao, J. (2003). Mathematical Statistics. Second edition. New York: Springer.

Surles, J. G. and Padgett, W. J. (2001). Inference for reliability and stress-strength for a scaled Burr type X distribution. Lifetime Data Analysis 7, 187–200.

Svensson, I. and Sjöstedt-de Luna, S. (2010). Asymptotic properties of a stochastic EM algorithm for mixtures with censored data. Journal of Statistical Planning and Inference
140, 111–127.

Szu, H. and Hartley, R. (1987). Fast simulated annealing. Physics Letters A 122, 157–162.

Tong, H. (1974). A note on the estimation of Pr{Y < X} in the exponential case. Technometrics 16, 625.

Tong, H. (1975). Errata. Technometrics 17, 395.

Tong, H. (1977). On the estimation of Pr{Y 6 X} for exponential families. IEEE Transactions on Reliability 26, 54–56.

Tregouet, D. A., Escolano, S., Tiret, L., Mallet, A. and Golmard, J. L. (2004). A new algorithm for haplotype-based association analysis: the stochastic-EM algorithm. Annals of Human Genetics 68, 165–177.

Xiang, Y., Gubian, S., Suomela, B. and Hoeng, J. (2013). Generalized simulated annealing for global optimization: the GenSA package. The R Journal 5, 13–28.

Xiang, Y., Sun, D. Y., Fan, W. and Gong, X. G. (1997). Generalized simulated annealing algorithm and its application to the Thomson model. Physics Letters A 233, 216–220.

Zhang, M., Ye, Z. and Xie, M. (2014). A stochastic EM algorithm for progressively censored data analysis. Quality Reliability Engineering International, 30, 711–722.