Share
تحليل تصميم العبور بفترتين ودراسة التاثير المتبقي == The Analysis of The Two Period Cross - Over Design And Study of The Residual Effect
Author name:
عقيل اكرم علي الزبيدي
Supervisor name:
لميعة باقر جواد الجواد
General topic:
Administration and Economics
Specific topic:
Statistics
Degree:
Doctorate
University:
University of Baghdad - Faculty Of Administration And Economics - Department Of Statistics
Language:
Arabic
University location:
Baghdad
First pages:
07T3858 - p.pdf
Abstract:
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
