ThanoonYThanoon -

Gender	MALE
Place of Work	Presidency
Position	Vice-Chancellor of President University for Administrative Affairs
Qualification	PhD
Speciality	Applied Statistics
Email	Thanoon.younis@ntu.edu.iq
Phone	00
Address	Iraq-Mosul, Iraq, Mosul, Iraq

Model comparison of Bayesian structural equation models with mixed ordered categorical and dichotomous data

Feb 24, 2017

Journal Journal of Statistics and Management Systems

Issue 1

Volume 20

The purpose of this paper is to describe the mixed variables (ordered categorical and dichotomous) in Bayesian structural equation models. Markov chain Monte Carlo simulation (MCMC) via Gibbs sampling method is applied for estimation the parameters. Statistical analyses, which include parameters estimation, standard error, higest posterior density and Devience information creterion for testing the prposed models, are discussed. Hidden continuous normal distribution with censoring is used to handle the problem of mixed variables (ordered categorical and dichotomous). Comparison between Bayesian linear and non-linear SEMs are discussed. The proposed models are illustrated by a case study for breast cancer patient’s which obtained from the hospital. Analyses are done by using WinBUGS program. The results showed that the results of non-linear Bayesian SEM is better than the results of linear Bayesian SEM.

Read Publication

ROW AND COLUMN MATRICES IN MULTIPLE CORRESPONDENCE ANALYSIS WITH ORDERED CATEGORICAL AND DICHOTOMOUS VARIABLES

Feb 10, 2016

Journal JURNAL TEKNOLOGI

DOI https://doi.org/10.11113/jt.v78.4077

Issue 2: February 2016

Volume 78

In multiple correspondence analysis, whenever the number of variables exceeds the number of observations, row matrix should be used, but if the number of variables is less than the number of observations column matrix is the suitable procedure to follow. One of the following matrices (rows, columns) leads to loss of information that can be found by the other method, therefore, this paper developed a proposal to overcome this problem, which is: to find a shortcut method allowing the use of the results of one matrix to obtain the results of the other matrix. Taking advantage of all information available, the phenomenon was studied. Some of these results are: Eigenvectors, factor loadings and factor scores based on ordered categorical and dichotomous data. This method is illustrated by using a real data set. Results were obtained by using Minitab program. As a result, it is possible to shortcut transformation between the results of row and column matrices depending on factor loadings and factor scores of the row and column matrices.

Read Publication

Bayesian Analysis of Linear and Nonlinear Latent Variable Models with Fixed Covariate and Ordered Categorical Data

Jan 1, 2016

Journal Pakistan Journal of Statistics and Operation Research

DOI https://doi.org/10.18187/pjsor.v12i1.952

Issue 12

Volume 1

In this paper, ordered categorical variables are used to compare between linear and nonlinear interactions of fixed covariate and latent variables Bayesian structural equation models. Gibbs sampling method is applied for estimation and model comparison. Hidden continuous normal distribution (censored normal distribution) is used to handle the problem of ordered categorical data. Statistical inferences, which involve estimation of parameters and their standard deviations, and residuals analyses for testing the selected model, are discussed. The proposed procedure is illustrated by a simulation data obtained from R program. Analysis are done by using OpenBUGS program.

Read Publication

Bayesian Analysis of Multiple Group Nonlinear Structural Equation Models with Ordered Categorical and Dichotomous Variables: A Survey

Dec 1, 2015

Journal Research Journal of Mathematical and Statistical Sciences

DOI https://doi.org/10.1063/1.4903693

Issue 12

Volume 3

This paper is designed to give a complete overview of the literature that is available, as it relates to application of the Bayesian analysis model to investigate multiple group nonlinear structural equation models, also known as SEMs, including those having ordered categorical, dichotomous and categorical-dichotomous mixed variables. It will also work to summarize Bayesian multiple group nonlinear SEMs with nonlinear covariate variables, and latent variables in the structural model and both linear covariant and latent variable sin the measurement models. More specifically, it will be suggested that using hidden continuous normal distribution, including both right and left censoring and truncation, and interval censoring and truncation, can improve the Bayesian approach to multiple group nonlinear structural equation models when solving problems using ordered categorical and dichotomous data.

Study of the Relationship between Dependent and Independent Variable Groups by Using Canonical Correlation Analysis with Application

Jul 20, 2015

Journal Modern Applied Science

publisher Canadian Center of Science and Education

DOI https://doi.org/10.1063/1.4903693

Issue 8

Volume 9

Canonical correlation analysis is used to study the relationship between two groups of variables (dependent and independent). Since each group represents the linear combination to a number of variables, canonical correlation analysis measures the relationship between these variables that maximally correlate with linear combinations of another subset of variables. Statistical analysis involves canonical correlation between two groups of variables, canonical variates, standard canonical variates, canonical factor loadings, canonical cross factor loadings for both groups. Test of significance of canonical correlation using Wilk's Lambda showed that the first and second canonical correlation was significant and the third and fourth canonical correlation were insignificant. This method is illustrated by using a real data set. Results obtained by using SPSS program.

Read Publication

GENERALIZED NONLINEAR CANONICAL CORRELATION ANALYSIS WITH ORDERED CATEGORICAL AND DICHOTOMOUS DATA

Jul 1, 2015

Journal JURNAL TEKNOLOGI

DOI https://doi.org/10.11113/jt.v75.3602

Issue 1

Volume 75

In this paper, ordered categorical and dichotomous data are used in generalized nonlinear canonical correlation analysis to study the relationship between two or more sets of variables. Statistical analyses involving generalized nonlinear canonical correlation analysis, component loadings, and object scores are discussed in this paper. The proposed procedure is illustrated using a real data set (patients with high blood pressure). Analyses are done using SPSS program. The component loadings graph of the three sets shows the relationship between the three sets and their impact on the data set of patients with high blood pressure. The centroid graph of the categories also shows the relationship between them.

Read Publication

Multiple factor analysis with continuous and dichotomous variables

Apr 1, 2014

DOI https://doi.org/10.1063/1.4903693

Issue December 04 2014

In this paper, continuous and dichotomous variables are used in multiple factor analysis method. When all variables within the same group are continuous, we use principal component analysis method in factor analysis, if all variables within the same group are dichotomous we use multiple correspondence analysis method in factor analysis. Statistical analyses, which involve Eigen roots, Eigen vectors, multiple factor loadings, correlation coefficient RV, contribution table, are discussed. The proposed procedure is illustrated by a lung cancer data consists of four groups "group of personal variables", "group of therapeutic variables", "group of nutritional variables", "group of genetic variables". Analysis are done by using XLSTAT program.

Read Publication