analysis of dependent repeated data by using univariate and multivariate approaches : a comparison study using simulation

العناوين الأخرى

تحليل البيانات المكررة المترابطة بأستخدام أساليب أحادية المتغير و متعدد المتغيرات : دراسة مقارنة باستخدام المحاكاة

مقدم أطروحة جامعية

Surji, Evar Lutf Allah Sadr al-Din

مشرف أطروحة جامعية

Kubbarah, Sami Dawud

أعضاء اللجنة

Abd Allah, Herish Umar

الجامعة

جامعة صلاح الدين

الكلية

كلية العلوم

القسم الأكاديمي

قسم الرياضيات

دولة الجامعة

العراق

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2010

الملخص الإنجليزي

Gabbara (1994) developed a general method to analyze a repeated measures model when the covariance matrix has certain structure (pattern).

Let 11 12  be the vector of observations on the individual.

It is assumed that the Yi are independently normally distributed with mean and structured covariance matrix   0.

Different procedures are used to test hypotheses about  i.

Certain Analysis of Variance (ANOVA) models with correlated data that have the same structure of the covariance matrix of Gabbara (1994) are used as applications.

For this, two cases are considered.

In case I, each individual receives the same treatments and hence has the same mean vector while in case II, different individuals receive different treatments and hence have different mean vectors.

A comparison has been made between univariate approaches including ANOVA and Repeated Measures Model (RMM), and multivariate approaches including Multivariate One Sample Model (MOSM) and Generalized Multivariate Linear Model (GMLM), where for the first case, ANOVA, RMM and MOSM procedures are used while for the second case, RMM and GMLM are used.

For this purpose, a simulation study has been conducted to generate data with covariance matrix that have the same structure as that of the covariance matrix of the RMM of Gabbara (1994).

In general, the results of type I error rates have shown that the RMM procedure is better, in comparison to the other used procedures.

التخصصات الرئيسية

الرياضيات

الموضوعات

• التحليل متعدد المتغيرات

عدد الصفحات

83

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-311124