新藥(原廠藥)的研發平均需花10~12年的時間和八百億美元，因此新藥的研發是非常耗時又需花費巨大金額，為賺取利潤使得新藥的價格是非常昂貴，於是政府為了降低藥物成本和價格，同意學名藥廠等到專利期滿，就可倣照原廠藥製造出具相同療效的藥(學名藥)，在1984年美國食品與藥物管理局(FDA)准許學名藥上市，只須証明學名藥與原廠藥是否具有平均生体相等性( bioequivalence)即可，而且學名藥不須長期的臨床試驗, 因此學名藥廠可以節省大量的金錢和研發時間。
　　目前，在評估平均生体相等性中，最大概似估計法已被廣泛地使用並接受，然而在這篇論文中，為達到統計上的不偏性和最小變異，於是採用最小變異不偏估計量(MVUE)來評估平均生体相等性。因此我們執行一個模擬研究，考慮不同交叉設計的參數組合和樣本數，以bias、mean square error(MSE)、經驗型誤(empirical size)、經驗檢定力(empirical power)和90%信賴係數來比較MLE和MVUE的優劣性。

　　The research and development of an innovative drug product in the average take 10-12 years and US $ 800 million dollars. Therefore, it is a costly, time-consuming, and highly risky endeavor. One way to reduce the drug cost is to introduce generic drugs after the patent of the innovative drugs expires. Currently, most regulatory agencies in the world only require evidence of average bioequivalence from in vivo bioequivalence trials to approve the generic drugs.
　　Currently, maximum likelihood estimator (MLE) is recommended for evaluation of average bioequivalence. However, we considered to adopt the minimum variance unbiased estimator (MVUE) to assess the average bioequivalence. We performed a simulation study to compare the bias, mean square error, empirical size, empirical power and 90% confidence coefficient between MLE and MVUE on the various combinations of parameters and sample size under 2 2 crossover design and higher-order crossover design.

1.Bradu, D. and Mundlak, Y (1970). ‘Estimation in
lognormal linear models’, Journal of the
American Statistical Association, 65, 198-211.
2.Hoyle, M.H. (1968). ‘The estimation of variances
after using a Gaussinating transformation’. Ann.
Math. Stat., 39, 1125-1143.
3.Jones, B. and Kenward, M.G. (1989). ‘Design and
Analysis of Crossover Trails. Chapman and Hall,
London, England.
4.Kershner, R.P. and Federer, W.T. (1981).
‘Two-treatment crossover design for estimating a
variety of effects’. J. Amer. Stat. Assoc., 76,
612-618.
5.Land, C.E. (1988). ‘Hypothesis tests and
interval estimates’, Chapter 4 in Crow, E.L. and
Shimizu, K. (eds.), Lognormal Distribution Marcel
Dekker, New York, 87-112.
6.Liu, J.P. and Chow, S.C (1992). ‘Design and
analysis of bioavailability and bioequivalence
studies’, Marcel Dekker, New Jersey.
7.Liu, J.P. and Weng, C.S (1992). ‘Estimation of
direct formulation effect under log-normal
distribution in bioavailability/bioequivalence
studies’, Statistics in Medicine, Vol 11,
881-896.
8.Mehran, F. (1973). ‘Variance of the MVUE for the
lognormal mean’. Journal of the American
Statistical Association, 68, 726-727.
9.Metzler, C.M. (1974). ‘Bioavailability: A
problem in equivalence’. Biometrics, 30,
309-317.
10.Neyman, J. and Scott, E. L (1960). ‘Correction
for bias introduced by a transformation of
variances’, Annals of Mathematical Statistics,
31, 643-655.
11.O’Brien, P.C. (1984). ‘Procedure for comparing
samples with multiple end points’. Biometrics,
40, 1079-1087.
12.Shimizu, K. (1988). ‘Point Estimation’,
Chapter 2 in Lognormal Distribution,ed. by Crow
and Shimizu, Marcel Dekker, New York, 27-86.
13.Smith, S.J. (1988). ‘Evaluating the efficiency
of the -distribution mean estimator’.
Biometrics, 44, 485-493.
14.U.S.FDA. Guidance for industry on
bioavailability and bioequivalence studies
for orally administered drug products-general
consideration. Center for Drug Evaluation and
Research, Food and Drug Administration,
Rockville, Maryland, 2003.
15.Westlake, W.J. (1976). ‘Symmetrical confidence
intervals for bioequivalence trails’.
Biometrics, 32, 741-744.
16.Westlake, W.J. (1981). ‘Bioequivalence
testing-A need to rethink’. (Reader reaction
response). Biometrics, 37, 591-593.