Classical Statistical Methods Research Articles

A Scoping Review of the Risk Factors Associated with Anaemia among Children Under Five Years in Sub-Saharan African Countries.

Background/Purpose: Globally, anaemia is a severe public health condition affecting over 24% of the world’s population. Children under five years old and pregnant women are the most vulnerable to this disease. This scoping review aimed to evaluate studies that used classical statistical regression methods on nationally representative health survey data to identify the individual socioeconomic, demographic and contextual risk factors associated with developing anaemia among children under five years of age in sub-Saharan Africa (SSA). Methods/Design: The reporting pattern followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) guidelines. The following databases were searched: MEDLINE, EMBASE (OVID platform), Web of Science, PUBMED, Cumulative Index to Nursing and Allied Health Literature (CINAHL), PsycINFO, Scopus, Cochrane library, African Journal of online (AJOL), Google Scholar and Measure DHS. Results: The review identified 20 relevant studies and the risk factors for anaemia were classified as child-related, parental/household-related and community- or area-related factors. The risk factors for anaemia identified included age, birth order, sex, comorbidities (such as fever, diarrhoea and acute respiratory infection), malnutrition or stunting, maternal education, maternal age, mother’s anaemia status, household wealth and place of residence. Conclusion: The outcome of this review is of significant value for health policy and planners to enable them to make informed decision that will correct any imbalances in anaemia across socioeconomic, demographic and contextual characteristics, with the view of making efficient distributions of health interventions.

TET2/IDH1/2/WT1 and NPM1 Mutations Influence the RUNX1 Expression Correlations in Acute Myeloid Leukemia.

Background and objectives: Mutational analysis has led to a better understanding of acute myeloid leukemia (AML) biology and to an improvement in clinical management. Some of the most important mutations that affect AML biology are represented by mutations in genes related to methylation, more specifically: TET2, IDH1, IDH2 and WT1. Because it has been shown in numerous studies that mutations in these genes lead to similar expression profiles and phenotypes in AML, we decided to assess if mutations in any of those genes interact with other genes important for AML. Materials and Methods: We downloaded the clinical data, mutational profile and expression profile from the TCGA LAML dataset via cBioPortal. Data were analyzed using classical statistical methods and functional enrichment analysis software represented by STRING and GOrilla. Results: The first step we took was to assess the 196 AML cases that had a mutational profile available and observe the mutations that overlapped with TET2/IDH1/2/WT1 mutations. We observed that RUNX1 mutations significantly overlap with TET2/IDH1/2/WT1 mutations. Because of this, we decided to further investigate the role of RUNX1 mutations in modulating the level of RUNX1 mRNA and observed that RUNX1 mutant cases presented higher levels of RUNX1 mRNA. Because there were only 16 cases of RUNX1 mutant samples and that mutations in this gene determined a change in mRNA expression, we further observed the correlation between RUNX1 and other mRNAs in subgroups regarding the presence of hypermethylating mutations and NPM1. Here, we observed that both TET2/IDH1/2/WT1 and NPM1 mutations increase the number of genes negatively correlated with RUNX1 and that these genes were significantly linked to myeloid activation. Conclusions: In the current study, we have shown that NPM1 and TET2/IDH1/2/WT1 mutations increase the number of negative correlations of RUNX1 with other transcripts involved in myeloid differentiation.

Characterizing genetically diverse blue grama [Bouteloua gracilis (Willd. ex Kunth) Lag. ex Griffiths] seed sources

The need for regionally adapted native grass seed sources for the northern Great Plains has resulted in the commercial release of a range of plant materials, including ecovars™ 1 . Ecovars™ are multisite composite populations developed to combine broad genetic diversity from a geographic region. The objective of this study was to determine whether morphological data could be used to distinguish between genetically diverse blue grama [Bouteloua gracilis (Willd. ex Kunth) Lag. ex Griffiths] seed sources through classical statistical methods. Entries included a Manitoba (MB) ecovar™, a USDA–NRCS released cultivar ‘Bad River’, and ecotypes from Wyoming and Minnesota. Vegetative and reproductive measurements and ratings were taken from a spaced-plant nursery during 2000–2001 in Carman, Manitoba, Canada. The results were analyzed using statistical techniques including: ANOVA, least significant difference, canonical discriminant analysis (CDA), and coefficients of variation. These techniques distinguished four genetically diverse seed sources from each other through CDA. As hypothesized, there was greater within-population genetic diversity for the MB ecovar™ and Wyoming and Minnesota ecotypes, compared with ‘Bad River’. Our results indicate that genetically diverse blue grama seed sources can be distinguished, based on phenotypic measurements.

Bioimage-Based Prediction of Protein Subcellular Location in Human Tissue with Ensemble Features and Deep Networks.

Prediction of protein subcellular location has currently become a hot topic because it has been proven to be useful for understanding both the disease mechanisms and novel drug design. With the rapid development of automated microscopic imaging technology in recent years, classification methods of bioimage-based protein subcellular location have attracted considerable attention for images can describe the protein distribution intuitively and in detail. In the current study, a prediction method of protein subcellular location was proposed based on multi-view image features that are extracted from three different views, including the four texture features of the original image, the global and local features of the protein extracted from the protein channel images after color segmentation, and the global features of DNA extracted from the DNA channel image. Finally, the extracted features were combined together to improve the performance of subcellular localization prediction. From the performance comparison of different combination features under the same classifier, the best ensemble features could be obtained. In this work, a classifier based on Stacked Auto-encoders and the random forest was also put forward. To improve the prediction results, the deep network was combined with the traditional statistical classification methods. Stringent cross-validation and independent validation tests on the benchmark dataset demonstrated the efficacy of the proposed method.

Dynamic measurement of the financial liquidity of Polish companies listed on the Warsaw Stock Exchange between 2002 and 2017: cross-sectoral comparative analysis - Part 2

The purpose of the present research was to analyse Polish listed companies in terms of liquidity management in 2002-2017 in a dynamie context. Evaluation of the dynamie model of corporate financial liquidity was carried out with the use of classical descriptive statistics tools and methods applied in sueh analyses. The eompanies were analysed in the new seetoral layout implemented by the Management Board of the Warsaw Stoek Exehange in January 2017. The study of the dynamie finaneial liquidity of enterprises on the basis of seleeted ratios eonsisted of two parts. In the first part of the study, a series of medians was determined for eaeh of the liquidity ratios in partieular seetors. Eaeh element of the series is assoeiated with the quarter from whieh the data eame. The ratios within sectors were then compared so that coexistence of changes over time, shifts in rela- tion to one another, or the lack of clear interdependencies could be observed. Unlike the first part, which covered the relationship between the different ratios within a sector, the second part focuses on cross-sectoral comparative analysis. Descriptive statistics based on quantiles were derived for data covering the entire period under consideration and for all enterprises in the sector. Half of the surveyed listed companies (i.e. all the ratio values between the first and third quartiles), were adopted as the central standard. Numerical values of standards for particular ratios and sectors are presented in a tabular form.

Comparative study on the time series forecasting of web traffic based on statistical model and Generative Adversarial model

We evaluated the accuracy of several classical statistical methods of Time series forecasting with ground truth dataset which was obtained from Kaggle web traffic forecasting competition hosted by Google. A novel way of seasonal, trend and cycle pattern decomposing method was used for the specific time series daily data. We proposed using the combination of four traditional methods to reduce the RMSE and thus achieved better forecasting accuracy. Results showed error rate was lowered down 10 to 20 percentage points. After studying the characteristics of the web traffic time series, we presented the Generative Adversarial Model (GAN) with Long-Short Term Memory (LSTM) as generator and deep Multi-Layer Perceptron (MLP) as discriminator to forecast the web traffic time series. The forecasting performances was compared among the traditional statistical methods and the deep generative adversarial network. We concluded from experiments there was no remarkable difference for this specific times series forecasting accuracy using these two kinds of methods.

Evaluation of the classical statistical, deterministic and geostatistical interpolation methods for estimating the groundwater level

Proper management of groundwater resources depends on identifying, modeling and predicting the level of groundwater in the plains for long-term planning and optimal use of the potential of water. To achieve continuous and integrated maps and predict unknown values, interpolation methods can be used. In this research, the classic statistical interpolation methods (including nearest neighbor, natural neighbor, moving average, and triangulation with linear interpolation), deterministic interpolation methods (including inverse distance to a power, radial basis functions and local polynomial) and geostatistical interpolation method (kriging) were used to estimate the groundwater level of Naqadeh plain in the northwest of Iran in the period of 2001–2016. The cross-validation technique was applied to evaluate the accuracy of the various methods, and three indices—the determination coefficient (R2), the root mean squared error and mean absolute error—were used to compare the interpolation methods. The results of geostatistical analysis indicated that the groundwater level is regionalized variable and there is a high spatial structure ratio between groundwater level data. The best-experimented variogram is a Gaussian model with a correlation coefficient of 0.981. According to the results of cross-validation method, the geostatistical interpolation method with the highest accuracy and the minimum error and the classic statistical interpolation methods with the least accuracy and maximum error were introduced as the optimal and inappropriate methods, respectively.

CUE: CpG impUtation ensemble for DNA methylation levels across the human methylation450 (HM450) and EPIC (HM850) BeadChip platforms

ABSTRACT DNA methylation at CpG dinucleotides is one of the most extensively studied epigenetic marks. With technological advancements, geneticists can profile DNA methylation with multiple reliable approaches. However, profiling platforms can differ substantially in the CpGs they assess, consequently hindering integrated analysis across platforms. Here, we present CpG impUtation Ensemble (CUE), which leverages multiple classical statistical and modern machine learning methods, to impute from the Illumina HumanMethylation450 (HM450) BeadChip to the Illumina HumanMethylationEPIC (HM850) BeadChip. Data were analysed from two population cohorts with methylation measured both by HM450 and HM850: the Extremely Low Gestational Age Newborns (ELGAN) study (n = 127, placenta) and the VA Boston Posttraumatic Stress Disorder (PTSD) genetics repository (n = 144, whole blood). Cross-validation results show that CUE achieves the lowest predicted root-mean-square error (RMSE) (0.026 in PTSD) and the highest accuracy (99.97% in PTSD) compared with five individual methods tested, including k-nearest-neighbours, logistic regression, penalized functional regression, random forest, and XGBoost. Finally, among all 339,033 HM850-only CpG sites shared between ELGAN and PTSD, CUE successfully imputed 289,604 (85.4%) sites, where success was defined as RMSE < 0.05 and accuracy >95% in PTSD. In summary, CUE is a valuable tool for imputing CpG methylation from the HM450 to HM850 platform.

Accuracies of Genomic Prediction for Growth Traits at Weaning and Yearling Ages in Yak.

Simple SummaryGenomic selection is a new technology in animal breeding after the selection according to the best linear unbiased prediction (BLUP) value and marker assisted selection. Genomic selection has gradually been used in practical applications over recent years following the advent of high-density single nucleotide polymorphism (SNP) chips for livestock and poultry. Yak are a critical species on the Qinghai–Tibet Plateau, which is of great significance to herders. The early selection of yak could save feeding costs and shorten the generation interval. In the present study, we estimated the accuracy of genomic prediction compared with different classical models for yak early growth traits. The results of cross-validation indicated that the average predictive accuracy ranged from 0.147 to 0.391. The average correlation coefficient between prediction and true phenotype was 0.4.Genomic selection is a promising breeding strategy that has been used in considerable numbers of breeding projects due to its highly accurate results. Yak are rare mammals that are remarkable because of their ability to survive in the extreme and harsh conditions predominantly at the so-called “roof of the world”—the Qinghai–Tibetan Plateau. In the current study, we conducted an exploration of the feasibility of genomic evaluation and compared the predictive accuracy of early growth traits with five different approaches. In total, four growth traits were measured in 354 yaks, including body weight, withers height, body length, and chest girth in two early stages of development (weaning and yearling). Genotyping was implemented using the Illumina BovineHD BeadChip. The predictive accuracy was calculated through five-fold cross-validation in five classical statistical methods including genomic best linear unbiased prediction (GBLUP) and four Bayesian methods. Body weights at 30 months in the same yak population were also measured to evaluate the prediction at 6 months. The results indicated that the predictive accuracy for the early growth traits of yak ranged from 0.147 to 0.391. Similar performance was found for the GBLUP and Bayesian methods for most growth traits. Among the Bayesian methods, Bayes B outperformed Bayes A in the majority of traits. The average correlation coefficient between the prediction at 6 months using different methods and observations at 30 months was 0.4. These results indicate that genomic prediction is feasible for early growth traits in yak. Considering that genomic selection is necessary in yak breeding projects, the present study provides promising reference for future applications.

Analytic Hierarchy Process–Based Selection Method for Airline Pilot Candidates

ABSTRACT Objective: In this study, a new selection method for airline pilot candidates is developed. Background: In a large majority of previous studies, parametric statistical classification methods have been commonly used for the pilot candidate selection process. These kind of methods require a representative sample with sufficient numbers of good and poor candidates to minimize the standard error associated with the regression weights and to ensure prediction accuracy. In this study, a new pilot candidate selection method has been developed, which can also be used in cases where there are not enough candidates or there is not a significant relationship between predictors. Method: The model was developed by using an analytic hierarchy process technique. A survey was conducted with airline pilots working at Turkish air carriers to determine the weights of the selection criteria considered in the model. The minimum success score of each criterion was obtained from the survey results. The analyzes are carried out using the Super Decisions program. Results: We found that the most important factor among the pilot candidate selection criteria is psychomotor coordination and operational abilities with a coefficient of 0.37306. Conclusion: The survey results revealed the importance of flight abilities and personality traits for pilots and it also found that lack of motivation is the most important problem among new airline pilots.

COMPARISONS OF SOME BIASED ESTIMATORS FOR LINEAR MEASUREMENT ERROR MODELS

Measurement errors are very often come upon in data analysis. Classical statistical methods become disadvantageous and ordinary least squares estimator of parameters turns into inconsistent and biased, in the existence of measurement errors in the data. Although some methods are used, the performances of them are not good enough in the presence of multicollinearity and measurement errors in the data, simultaneously. That’s why researchers have been inquiring about the estimation of the parameters if the measurement error models have multicollinearity, lately. Especially, biased estimation techniques have been researched in the existence of multicollinearity for measurement error models recently. In this paper, the ridge and Liu estimation approaches to the measurement error models in the existence of multicollinearity are investigated. The comparisons of the biased estimators’ performances are analyzed theoretically and numerically.

Information loss resulting from statistical disclosure control of output data

The most important methods of assessing information loss caused by statistical disclosure control (SDC) are presented in the paper. The aim of SDC is to protect an individual against identification or obtaining any sensitive information relating to them by anyone unauthorised. The application of methods based either on the concealment of specific data or on their perturbation results in information loss, which affects the quality of output data, including the distributions of variables, the forms of relationships between them, or any estimations. The aim of this paper is to perform a critical analysis of the strengths and weaknesses of the particular types of methods of assessing information loss resulting from SDC. Moreover, some novel ideas on how to obtain effective and well-interpretable measures are proposed, including an innovative way of using a cyclometric function (arcus tangent) to determine the deviation of values from the original ones, as a result of SDC. Additionally, the inverse correlation matrix was applied in order to assess the influence of SDC on the strength of relationships between variables. The first presented method allows obtaining effective and well- -interpretable measures, while the other makes it possible to fully use the potential of the mutual relationships between variables (including the ones difficult to detect by means of classical statistical methods) for a better analysis of the consequences of SDC. Among other findings, the empirical verification of the utility of the suggested methods confirmed the superiority of the cyclometric function in measuring the distance between the curved deviations and the original data, and also heighlighted the need for a skilful correction of its flattening when large value arguments occur.

Machine Learning for Plant Breeding and Biotechnology

Classical univariate and multivariate statistics are the most common methods used for data analysis in plant breeding and biotechnology studies. Evaluation of genetic diversity, classification of plant genotypes, analysis of yield components, yield stability analysis, assessment of biotic and abiotic stresses, prediction of parental combinations in hybrid breeding programs, and analysis of in vitro-based biotechnological experiments are mainly performed by classical statistical methods. Despite successful applications, these classical statistical methods have low efficiency in analyzing data obtained from plant studies, as the genotype, environment, and their interaction (G × E) result in nondeterministic and nonlinear nature of plant characteristics. Large-scale data flow, including phenomics, metabolomics, genomics, and big data, must be analyzed for efficient interpretation of results affected by G × E. Nonlinear nonparametric machine learning techniques are more efficient than classical statistical models in handling large amounts of complex and nondeterministic information with “multiple-independent variables versus multiple-dependent variables” nature. Neural networks, partial least square regression, random forest, and support vector machines are some of the most fascinating machine learning models that have been widely applied to analyze nonlinear and complex data in both classical plant breeding and in vitro-based biotechnological studies. High interpretive power of machine learning algorithms has made them popular in the analysis of plant complex multifactorial characteristics. The classification of different plant genotypes with morphological and molecular markers, modeling and predicting important quantitative characteristics of plants, the interpretation of complex and nonlinear relationships of plant characteristics, and predicting and optimizing of in vitro breeding methods are the examples of applications of machine learning in conventional plant breeding and in vitro-based biotechnological studies. Precision agriculture is possible through accurate measurement of plant characteristics using imaging techniques and then efficient analysis of reliable extracted data using machine learning algorithms. Perfect interpretation of high-throughput phenotyping data is applicable through coupled machine learning-image processing. Some applied and potentially applicable capabilities of machine learning techniques in conventional and in vitro-based plant breeding studies have been discussed in this overview. Discussions are of great value for future studies and could inspire researchers to apply machine learning in new layers of plant breeding.

Risks of Islamic Banking Implementation (Through the Example of the Russian Federation)

The presented study examines the potential risks of Islamic banking during its implementation (through the example of the Russian Federation).Aim. The study aims to identify the potential risks of Islamic banking in Russia.Tasks. The author considers the main differences between traditional and Islamic banks, systematizes the types of risks in Islamic banks, analyzes Sharia and economic risks, and examines the potential risks of Islamic financial institutions through the example of the Amal Financial House.Methods. This study uses the dialectical method, methods of statistical research, classification, systems and structural analysis, graphical analysis, generalization and systematization, scientific methods of cognition (observation, comparison, analysis, and synthesis).Results. The main differences between traditional and Islamic banks and the types of risks in Islamic banks are systematized. The system of Sharia control is considered through the example of an Is lamic financial organization (Islamic banking). Through the example of the Amal Financial House, the major risks are identified — for Islamic banking in financing using such Islamic financial instruments as murabaha and ijara, and for investors using mudaraba.Conclusions. Systematization of the main differences between Islamic and traditional banks shows that operations are based on a real asset to exclude contradictions with Sharia (Islamic canon law). The author distinguishes between two types of risk in Islamic banks: economic and Sharia. Sharia control is divided into internal and external (through the example of the Amal Financial House), which ensure the functioning of Islamic banking to prevent risk in unauthorized transactions within the framework of Islam. The author analyzes the activities of the Amal Financial House in 2015-2019 related to financing using such Islamic financial instruments as murabaha and ijara. The return on investment products shows positive dynamics, i.e. investors receive a steady income, even though there is a risk of losing the invested funds.

The usefulness of robust multivariate methods: A case study with the menu items of a fast food restaurant chain

Multivariate statistical methods have been playing an important role in statistics and data analysis for a very long time. Nowadays, with the increase in the amounts of data collected every day in many disciplines, and with the raise of data science, machine learning and applied statistics, that role is even more important. Two of the most widely used multivariate statistical methods are cluster analysis and principal component analysis. These, similarly to many other models and algorithms, are adequate when the data satisfies certain assumptions. However, when the distribution of the data is not normal and/or it shows heavy tails and outlying observations, the classic models and algorithms might produce erroneous conclusions. Robust statistical methods such as algorithms for robust cluster analysis and for robust principal component analysis are of great usefulness when analyzing contaminated data with outlying observations. In this paper we consider a data set containing the products available in a fast food restaurant chain together with their respective nutritional information, and discuss the usefulness of robust statistical methods for classification, clustering and data visualization.

Second Order Statistical Method for Breast Thermal Images Classification

Second Order Statistical Method for Breast Thermal Images Classification

Working memory, executive functions, and emotional intelligence in second language writing

The current study employed an audiovisual blended-emotion stimulus to explore whether particular aspects of working memory (WM) and trait emotional intelligence (TEI) are significant determinants of linguistic complexity, accuracy, and fluency in second language (L2) writing. Fifty-nine L2 learners of Spanish completed a visuospatial WM task, four executive functioning tasks assessing inhibition, shifting, and updating abilities, a self-reported measure of TEI, and a writing task based on an animated short film that required participants to describe the plot of the film and express how they felt about it. Classical and Bayesian statistical methods were used to analyse the data. Updating ability proved to be the best variable to predict both subordination density and linguistic accuracy in L2 writing. Individual differences in controlled search and retrieval processes seem to account for these results. On the other hand, TEI was not linked to any of the L2 writing performance measures but did correlate negatively with updating ability. Furthermore, a trade-off was observed between fluency and lexical complexity, suggesting that paying attention to lexical choices to convey (emotional) meaning might slow down production speed.

Proof-of-Concept and Feasibility Study to Evaluate the Effect of β-Glucan on Protective Qi Deficiency in Adults.

To preliminarily explore the potential effect of β-glucan on Chinese medicine (CM) concept protective qi deficiency (PQD), and the methodology for future definitive studies. To have a standardized assessment of PQD, a list of 13 potentially PQD-relevant parameters were firstly created, each with defined quantitative or categorial scales. Using the data from 37 participants with (21 cases) or without (16 cases) PQD, multivariate logistic modeling was conducted to create a preliminary diagnostic PQD risk score. Subsequently, 21 participants diagnosed with PQD were treated with β-glucan in a dose of 200 mg/day for 8 weeks. Data were collected for trial acceptability measures (rate of recruitment, withdrawal, and compliance), and the participants were assessed for PQD status at baseline and every 2 weeks thereafter. The preliminary logistic model consisted of 3 parameters (low voice and apathy, aversion to wind and cold, and Cun pulse). The resulting risk score demonstrated a degree of PQD-predicting accuracy that, as evaluated by statistical (discrimination and classification) methods, was higher than those obtained from any of the individual candidate parameters. The 21 PQD participants treated with β-glucan demonstrated good receptibility and a time-dependent improvement in PQD status as evidenced by the decrease of PQD participant to 9.5% at the end of study. This study demonstrated the effect of proof-of-concept of β-glucan on improving PQD and the proof-of-concept of a multivariate-model-derived diagnostic PQD risk score. It also indicated feasibility for future definitive studies. Studies like this embody an innovative approach that uses therapies derived from the mainstream biomedicine to enrich therapeutics guided by CM principle. (Trial registration No. NCT03829228).

Modern Kredi Sınıflandırma Çalışmaları ve Metasezgisel Algoritma Uygulamaları: Sistematik Bir Derleme

Number of proposed advanced analysis methods, which try to successfully predict if applicants are going to default in credit applications show an increasing pattern, especially after the Global Financial Crisis. Alternative to conventional statistical classification methods, machine learning methods arrive on the scene; they have capability to reveal information from the data independently from constraints and assumptions. Along with machine learning methods, metaheuristic algorithms that substantially improves classification performances take part in studies. Combined usages of learning methods and metaheuristic algorithms aim to benefit from the contemporary data storage and process capacities at the highest level and greatly contribute to credit risk assessment field. In this review study, credit classification studies that adopt metaheuristic algorithms in the analyses are examined with a systematic process, for the period after&nbsp;2000. By forming a general framework, classification methods, metaheuristic algorithm implementations, algorithms’ intended uses and performance assessment criteria are addressed. Examination showed that there is a growing interest, nevertheless method preferences are concentrated over a limited option. It is necessary to incorporate more novel metaheuristics and/or hybrid and combined usages to the studies. It is possible to say that progressive works parallel to the developments in computer and mathematical sciences will continuously contribute to the literature.

Predicting the Session of an P2P IPTV User through Support Vector Regression (SVR)

Scalability and ease of implementation make Peer-to-Peer (P2P) infrastructure an attractive option for live video streaming. Peer end-users or peers in these networks have extremely complex features and exhibit unpredictable behavior, i.e. any peer may join or exit the network without prior notice. Peers' dynamics is considered one of the key problems impacting the Quality of Service (QoS) of the P2P based IPTV services. Since, peer dynamics results in video disruption to consumer peers, for smooth video distribution, stable peer identification and selection is essential. Many research works have been conducted on stable peer identification using classical statistical methods. In this paper, a model based on machine learning is proposed in order to predict the length of a user session on entering the network. This prediction can be utilized in topology management such as offloading the departing peer before its exit. Consequently, this will help peers to select stable provider peers, which are the ones with longer session duration. Furthermore, it will also enable service providers to identify stable peers in a live video streaming network. Results indicate that the SVR based model performance is superior to an existing Bayesian network model.