ISSN 1842-4562
Member of DOAJ
ISSUES
Volume 17 - 2022
Volume 16 - 2021
Volume 15 - 2020
Volume 14 - 2019
Volume 13 - 2018
Volume 12 - 2017
Volume 11 - 2016
Volume 10 - 2015
Volume 9 - 2014
Volume 8 - 2013
Volume 7 - 2012
Volume 6 - 2011
Volume 5 - 2010
Volume 4 - 2009
Volume 3 - 2008
Volume 2 - 2007
Volume 1 - 2006

Journal Home > Volume 11, Issue 4 - December 30, 2016

JAQM Volume 11, Issue 4 - December 30, 2016




Contents


Conjoint Representation of the Main Effects and Interaction Term in Multiple Non Symmetrical Correspondence Analysis
Antonello D’AMBRA, Anna CRISCI, Enrica SEPE

The main effects represent the change in the response variables due to the change on the level/categories of the predictor variables, considering the adding effects of them. By contrast the interaction effect represents the combined effect of predictor variables on the response variable. In particular, there is an interaction between two predictor variables when the effect of one predictor variable varies as the levels/categories of the other predictor vary. If the interaction is not significant, it is possible to examine the main effects. Instead, if the interaction is statistically significant and of strong entity, then, it is not useful to consider the main effects. As the matter of fact, asserting that two predictor variables interact is the same as affirming that the two variables do not have separate effects. Moreover, in this paper we suggest a procedure of the simultaneous representation of the main effect and interaction term obtained by means the decomposition of tau Gray Williams. To identify a category which is statistically significant, the confidence ellipses for a Multiple Non-Symmetric Correspondence Analysis will be shown.

Does Economic Status Matter for the Regional Variation of Malnutrition-Related Diabetes in Romania? Temporal Clustering and Spatial Analyses
Elena DRUICA, Zizi GOSCHIN

Among the different types of diabetes, in Romania the malnutrition-related diabetes displays the highest territorial inequality. In this paper, we combined two types of statistical tools, temporal clustering and spatial analysis, to find some relevant patterns in its territorial distribution. Firstly we conducted a time series clustering for the 41 counties and Bucharest Municipality, over 2007-2014, based on CORT and ACF dissimilarity distances and choose four clusters in each case. Within each cluster the evolution of malnutrition diabetes is similar. The clusters were then included as dummy variables in a spatial model testing the determinants of malnutrition-related diabetes incidence at county level. Malnutrition-related diabetes is a disease that might be linked to the economic status, therefore GDP per capita and average wage have been tested and found significant as factors of influence in various model specifications. The dummies representing the temporal clusters are also significant determinants of the regional incidence of malnutrition-related diabetes in Romania. We found that when introducing the cluster dummies in the spatial model, it becomes less appropriate than classic OLS regression, which suggests that temporal clusters were able to capture the spatial dependence in our data. The contribution of our work is three folded. First, we applied time series clustering in R and in doing so we added a real – data application to this scarce stream of literature. Secondly, we combined two techniques relatively new in Romanian data: spatial analysis and time series clustering. Last, but not least, we discussed the malnutrition – related diabetes, mellitus, as a possible proxy of poverty, and tried to advocate our claim by relating this disease’s territorial distribution with some economic variables.

An Empirical Assessment of Over-Dispersion in Zero Inflated Poisson and Conway-Maxwell Poisson Probability Distributions
R. M. SAKIA

A zero inflated probability model arises when probability mass at point zero exceeds the one allowed under the standard parametric family of discrete distribution. Ignoring zero inflation can as a consequence, result in biasedness in the estimated parameters and standard errors as well as causing over-dispersion when fitting a discrete generalized linear model. In the case of a Poisson distribution, the equi-dispersion property no longer holds rendering the model not suitable for analysis. Variants of the Poisson distribution model have been proposed and studied. Among these are the Conway-Maxwell Poisson (COM-Poisson) and the Zero-inflated Poisson (ZI-Poisson). In this paper, we empirically evaluate the equi-dispersion property when the two models are fitted to some data. It is observed that whereas the parameter transformed COM-Poisson model (Arua and Sakia, 2015) induced a near- perfect equi-dispersion to the data, the Zero inflated model still exhibited some over-dispersion despite a substantial reduction of the zero inflation factor.

Fraud and Error as Object of Financial Audit
Ana-Maria NICULAE

This study analyzes the influence of the auditors in the abscission of fraudulent activities in Romania, the connection between the relation that might become between the auditor and the audited company, and also the way the quality of the audit depends on the audit report. The reason of initiation this research is represented by the importance of the existence of audit missions and of the final reports drafted by the auditors to reduce the fraud phenomenon which is the result of many gaps in the national economy and also in the international one. The research tries to highlight the importance the integrity of executive lead has in developing the normal activity of a company and in the remove of fraudulent activities. Studying this relation association of the auditor and client it is necessary to prove if this relation has a positive or negative influence over the auditor’s integrity. In the content of this study was used a sample made of 44 companies listed at Bucharest Stock Exchange from the Regulation Market and AeRO Market who owns various Standard actions, more specific, Aero Standard. Colleting the necessary information for the aim of the final results, was made of the financial situations of the company, form the annual report and from the audit report. Testing the 3 hypotheses from this study was made due to regression models. As a result of the analyze it has proven that in 2012, the lack of the executive leading integrity, represents an element which determines fraud. Between the existence of a social relation with the auditor and the leading client society and the impaired of the auditor’s integrity, it has been proven a positive association in 2012 and 2014, and a negative association in 2013. The research result over the auditor’s quality had prove that the audit report does not have an influence over the audit quality.

Application of Univariate Frailty Models in Modeling Survival Data with a Cured Fraction
Gurprit GROVER, Komal GOEL, Divya SETH

Cure models have been widely used in biomedical research and clinical trials to analyze survival data, when the study population is a mixture of susceptible and non-susceptible individuals. These models assume that individuals experiencing the event of interest are homogeneous in risk. However, there remains a degree of heterogeneity induced by unobservable risk factors, which may lead to distorted results. In order to model that unobserved heterogeneity in risk among susceptible in addition to incorporation of a cured component, frailty cure models can be the most appropriate choice. This article aims to estimate the cure fraction with frailty along with the investigation of prognostic factors that influence the survival of cancer patients. A retrospective data of 285 melanoma cancer patients is analyzed without frailty to Cox PH and AFT models and with frailty to only AFT model. Model selection is ascertained using Akaike Information Criterion. Estimates are obtained by using STATA software.

Dynamics of Banks’ Market Concentration in USA
Elena Cristina CANEPA, Bogdan CANEPA

Banks’ market concentration has considerably changed over the years preceding the 2008 financial crisis and also as a result of the subsequent defaults, bailouts and acquisitions. The connection between banks’ market concentration and the stability of the banking system needs to be thoroughly studied, in order to prevent future crisis. As a starting point, an analysis of the banks’ market concentration over the years can be helpful for both regulators and legislators. In this paper we study the dynamics of the U.S. banks’ total assets between 1976 and 2010. We fit different theoretical distributions to the data, examining also the goodness of fit and providing the estimated parameters. We find that the Kolmogorov-Smirnov test rejects the hypothesis that the entire data from March 1983 or March 1998 fits a generalized Pareto distribution, a normal distribution or a log-normal distribution. For March 1998, March 2003 and March 2008 we also perform a cross- sectional analysis, since the parameters of the fitted distribution change by size. The Kolmogorov-Smirnov test does not reject the hypothesis that the modified samples fit a generalized Pareto distribution, when the extreme deciles are taken out.

Factorial Experimental Design with Non-Normally Distributed Responses: an Application in Agriculture
Bahr Kadhim MOHAMMED, Ali Sadig Mohommed BAGER, Meshal Harbi ODAH

In this paper, we study the factors that have the greatest impact on wheat production in Iraq, by using a factorial experimental design, which has three factors and each factor has two levels. These factors are described as: the type of wheat, agriculture date, and types of used fertilizer. In this study, we use two methods: first method is the variance of analysis (ANOVA) and the second is the Log-transformation method. The random error is distributed as non-normal distribution due to the fact that response variable is distributed as non-normal distribution. The comparison between the two methods is based on the effects of the main treatments, the interactions between factors, and expected of length of confidence intervals [E(LOCI)] for the response variable. We used R packages (vcd and MASS) for the application.