مشاركة
تحليل تصميم العبور بفترتين ودراسة التاثير المتبقي == The Analysis of The Two Period Cross - Over Design And Study of The Residual Effect
اسم المؤلف:
عقيل اكرم علي الزبيدي
اسم المشرف:
لميعة باقر جواد الجواد
الموضوع العام:
الادارة والاقتصاد
السنة:
2009
الموضوع الدقيق:
الاحصاء
الدرجة:
دكتوراه
الجامعة:
جامعة بغداد
اللغة:
العربية
مكان الجامعة:
بغداد
الصفحات الاولى:
07T3858 - p.pdf
المستخلص:
في تجارب العبور كل وحدة تجريبية تستقبل اثنين من المعالجات او اكثر خلال الزمن.في ابسط الحالات تستقبل اثنين في المعالجات الوحدة التجريبية تعطي اولا واحدة من المعالجات ثم (تعبر) لتستقبل المعالجة الثانية.تصميم العبور بفترتين كثيرا ما يستخدم في التجارب الطبية | In crossover trials each experimental unit receives two or more treatments through time ; in the simplest case of two treatments , the subject is first given one of the treatments and then crosses over to the other treatment.The tow - period cross - over design is rather often used in medical trials ,especially with chronic diseases or in drug experiments with volunteers.With this design the error variance is reduced by applying both treatments to the same subjects ;and to exclude time (period)effects ,one group of subjects receive the treatments in the sequence AB and the other group receives the treatments in the sequence BA The disadvantage of this design is that in the second period there may be a residual effect of the treatments given in the first period.When the assumption of normal distribution and homogenous of variances are hold the perfect procedure to analysis (TPCOD) is F test and t test but when the assumptions are not hold or the data are pure ordinal like scores , since sums or differences of such data or not reasonable, therefore, A nonparametric model for cross over design is best.We select three experiments to compare between the parametric and nonparametric procedures.we proposal a nonparametric procedure to estimate and test the (TPCOD) that is adopt the nonparametric procedure against the parametric in the case of violation of assumptions
