如何推論工業產品之壽命分配以及提供精準的可靠度資訊，
一直是生產製造商十分重視的課題。韋伯與對數常態分配為目前工業界所廣泛使用的壽命分配，
但是當此兩種分配皆不足以來描述產品壽命分配
時，可考慮更一般化的對數-位置-尺度-形狀分配，如廣義 F 分配來描述之。
廣義 F (General F, GF)分配可包含廣泛的壽命分配 (韋伯，對數常態及廣義加瑪 (Gamma) 分配皆為其特例)，
本文首先探討一般化對數-位置-尺度-形狀模型之統計性質，
並探討誤判下產品可靠資訊 (如平均失效時間， p-th 百分位數) 之漸近分配。
接著以廣義加瑪分配為例，說明若被誤判成韋伯或對數常態分配時，如何在大樣本
理論下，以漸近均方差為準則，來找出最佳臨界值，進而挑選出穩健的模型。

[1] Farewell, V.T. and Prentice R.L. (1977). “A study of distribution shape
in life testing,” Technometrics, 19 , 69-75
[2] Hager, H. W. and Bain, L. J. (1970). “Inferential procedures for the
generalized gamma distribution,” Journal of the American Statistical
Association, 65 ,1601-9.
[3] Kalbfleisch, J.D., and Prentice R.L. (2002). The Statistical Analysis of
Failure Time Data, John Wiley and Sons, New York.
[4] Lawless, J. F. (2003). Statistical Models and Methods for Lifetime Data,
John Wiley and Sons, New York.
[5] Meeker, W. Q., and Escobar, L. A.(1998). Statistical Methods for Relia-
bility Data Analysis, John Wiley and Sons, New York.
[6] Pascual, F. G., and Montepiedra, G. (2005). “Lognormal and Weibull
Accelerated Life Test Plans Under Distribution Misspecification,” IEEE
Transactions on Reliability, 54, 43-52. 48, 11-25.
[7] Prentice R.L. (1974). ”A log gamma model and its maximum likelihood
estimation,” Biometrika, 61 , 539-44.
[8] Prentice R.L. (1975). “Discrimination among some parametric models,”
Biometrika, 62 , 607-14.
[9] White, H. (1982). “Maximum Likelihood Estimation of Misspecified
Models,” Econometrica, 50, 1-25.
75
[10] White, H. (1994). Estimation, Inference, and Specification Analysis,
John Wiley and Sons, New York.