國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
61
第五章、結論與建議
第一節、 結論
本研究使用貝氏方法估計 Huang(2004)的隨機作答模式之參數,能解決當
1 2
( , ˆ ˆ )
的值較極端 MLE ˆ
超過 0 或 1,而產生不合理估計值的情形,使用貝氏方 法估計能幫助社會統計學家得到合理的估計值,以作社會學相關參數的推論之用。其次,本研究驗證當事前資訊(prior information)提供
的事前期望值介於 0.05~0.33 時,貝氏估計量
*的估計效率高於 MLE ˆ
。第二節、 建議
本研究建議未來可再進行的研究,有兩大方向,分別為「貝氏估計方法」與
「再考慮非敏感性的群體有不完全誠實作答之情況」。以下為「貝氏估計方法」
可在著眼的項目:
1. 採用本研究的假設,討論當
的事前期望值大於 0.33 時,貝氏估計量
*與MLE ˆ
效率之高低。2. 若當
的事前分配不同於本研究之假設時,如:
與
不獨立時或是其邊際事前分配不為 beta 分配時,則貝氏估計量與貝氏風險為何?而貝氏估計量
*的估計效率與 MLE ˆ
比較的結果又為何?3. 直接假設
的事前分配從
下手推導
的事後分配,其式子由n r
1+1個機 率密度函數所組成,且貝氏估計量由n r
1+1個 beta 分配的期望值相加,當樣本數
n
很大時,計算就非常耗時。可再研究其他的方法,以求得簡化之近似‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
62
式。
Warner(1965)在其研究中假設使用直接詢問法時,不屬於敏感性群體的受訪 者會有不誠實作答的情況,但在其隨機作答模式中卻沒有考慮,Huang(2004)亦 無討論。若要再考慮非敏感性的群體有不完全誠實作答之情況,且想要對其不誠
實作答率b做估計時,則可研究新的隨機作答模式或是改良既有之隨機作答模式,
以同時取得、與b之估計量。
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
63
參考文獻
Abul-Ela, A. L. A., Greenberg, B. G., and Horvitz, D. G. (1967). “A
Multi-Proportional Randomized Response Model,” Journal of the American
Statistical Association, 62, 990-1008.
Bar-Lev, S. K., Bobovich, E., and Boukai, B. (2003). “A Common Conjugate Prior Structure for Several Randomized Response Models,” TEST, 12, 101-113.
Barabesi, L., & Marcheselli, M. (2006). “A Generalization of Huang’s Randomized Response Procedure for the Estimation of Population Proportion and Sensitivity Level.” Metron, vol. LXIV, n. 2, pp. 145-159.
Chang, H. J., and Huang, K. C. (2001). “Estimation of Proportion and Sensitivity of a Qualitative Character,” Metrika, 53, 269-280.
Chang, H. J., and Liang, D. H. (1996a). “A Two-Stage Unrelated Randomized Response Procedure for,” Australian journal of statistics, 38, 43-51.
Chang, H. J., and Liang, D. H. (1996b). “A Randomized Response Procedure for Two-Unrelated Sensitive Questions,” Journal of Information & Optimization
Sciences, 17, 185-198.
Chaubey, Y.,and Li, W. (1995). “Comparison between Maximum Likelihood and Bayes Methods for Estimation of Binominal Probability with Sample Compositing,”
Journal of Official Statistics, 11,379-390.
Chaudhuri, A., Mukerjee, R. (1988). Randomized Response: Theory and Techniques.
Marcel Dekker, New York.
Christofides, T. C. (2003). “A Generalized Randomized Response Technique,”
Metrika, 57, 195-200.
Christofides, T. C. (2005). “Randomized Response in Stratified Sampling,” Journal of
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
64
Statistical Planning and Inference, 128, 303-310.
Fidler, D. S., and Kleinknecht, R. E. (1977). “Randomized Response Versus Direct Questioning: Two Data-Collection Methods for Sensitive Information,”
Psychological Bulletin, 84, 1045-1049.
Greenberg, B. G., Abul-Ela, A. L. A., Simmons, W. R., and Horvitz, D. G. (1969).
“The Unrelated Question Randomized Response Model: Theoretical Framework,”
Journal of American Statistical Association, 64, 520-539.
Greenberg, B. G., Kuebler, R. R., Jr., Abernathy, J. R., and Horvitz, D. G. (1971).
“Application of the Randomized Response Technique in Obtaining Quantitative Data,” Journal of American Statistical Association, 66, 243-250.
Huang, K. C. (2004). “A Survey Technique for Estimating the Proportion and
Sensitivity in a Dichotomous Finite Population,” Statistica Neerlandica, 58, 75-82.
Kim, J. M., Tebbs J. M., and An S. W. (2006). “Extensions of Mangat’s Randomized Response Model,” Journal of Statistical Planning and Inference, 136, 1554-1567.
Kim, J. M., and Warde, W. D. (2004). “A Stratified Warner’s Randomized Response Model,” Journal of Statistical Planning and Inference, 120, 155-165.
Kuk, A. Y. C. (1990). “Asking Sensitive Questions Indirectly,” Biometrika, 77, 436-438.
Mangat, N. S., and Singh, R. (1990). “An Alternative Randomized Response Procedure,” Biometrika, 77, 439-442.
Mangat, N. S. (1994). “An Improved Randomized Response Strategy,” Journal of the
Royal Statistical Society: Series B, 1, 93-95.
Migon, H. S., and Tachibana, V. M. (1997). “Bayesian Approximations in
Randomized Response Model,” Computational Statistics & Data Analysis, 24, 401-409.
Pitz, G. F. (1980). “Bayesian Analysis of Random Response Models,” Psychological
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
65
Bulletin, 87, 209-212.
Singh, J. (1976). “Randomized Response a Method for Sensitive Surveys.” In
Proceedings of the Social Statistics Section, p. 722. American Statistical
Association.
Winkler, R. L., and Franklin, L. A. (1979). “Warner’s Randomized Response Model:
A Bayesian Approach,” Journal of the American Statistical Association, 74, 207-214.
Warner, S. L. (1965). “Randomized Response : A Survey Technique for Estimating Evasive Answer Bias,” Journal of the American Statistical Association, 60, 63-69.
王智立、蔡宛容,2007。應用一般化 Greenberg 無關聯隨機化作答模式於敏感問 題之研究,中國統計學報,第 45 卷,頁 189-205。
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
66