DIVESTITURE POST MERGERS AND ACQUISITIONS IN INDIA – REASONS AND MODEL TO PREDICT

Background: Failure to control diabetes on time leads to irreparable complications in other parts of body such as heart, kidneys, eyes, and etc. As those most susceptible to the disease are pre-diabetic individuals, this study aims to identify the Determinants for diabetes in people with pre-diabetes employing two advanced statistical methods of logistic regression and discriminant analysis. Methods: Data were collected from 17 health centers in the rural area of Kermanshah city. 100 diabetic patients and 100 pre-diabetes persons (controls) were enrolled. Demographic data, body mass index, fasting blood sugar (FBS), oral glucose tolerance test (OGTT), blood pressure, blood lipid levels and daily activity in two separate forms by the staff of Disease control from health records were collected. Logistic regression and discriminant analysis to identify Determinants and ROC curve were used to compare the predictive power of the models. Results: The predictive power of logistic regression and discriminant analysis were 0.884 and 0.80 respectively. Sex (P= 0.027) and FBS (P<0.001) in logistic regression and age (P=0.014), FBS (P<0.001) and OGTT (P<0.001) were significant in the discriminant analysis. Logistic regression model was more sensitive (79%). Conclusion: In this study, logistic regression was more powerful in the separation of patients from pre-diabetic. In communities that have high affinity between case and control groups, identifying differences needs stronger methods. Thus, using these methods recommended in medical studies.

This paper focuses on the problems of analysing growth and success of a firm. The data consists of 348 growing SMEs. 75 of them were growing rapidly and highly successful (HGS). The methods include discriminant, regression and logistic regression analysis. In regression models, growth and success seem to be inversely related. Regression analysis is not an appropriate way to analyse HGS firms. Logistic regression and discriminant analysis should consider a priori probabilities to produce reliable results. The use of different methods depends on the design of the study, the characteristics of the data and the validity of the research questions. A robust analysis presupposes refinement of the data, accurate model specification, and competent interpretation of the results. Future research should focus on periods longer than 3–5 years and take into account uncertainty and discontinuity of variables. A qualitative approach where behavioural and management issues will be included is needed.

Logistic regression modeling and the number of events per variable: selection bias dominates

SECURITY PRICES, RISK, AND MAXIMAL GAINS FROM DIVERSIFICATION*

Neural networks can be applied to many predictive data mining applications due to their power, flexibility and relatively easy operations. Predictive neural networks are very useful for applications where the underlying process is complex, such as in classification using a mix of nominal and ratio level variables and for predictive validity based on classification modelling. A neural network can approximate a wide range of statistical models without requiring the researcher to hypothesize in advance certain relationships between the dependent and independent variables. The two major applications are multilayer perceptron (MLP) and radial basis function (RBF) procedures. In contrast to MLP networks, in the RBS networks it is only the output units that have a bias term. Discriminant analysis (or discriminant function analysis) based on classification modelling is applied to classify cases into the values of a categorical dependent variable, usually a dichotomy. Logistic regression is useful for situations in which you want to be able to predict the presence or absence of a characteristic or outcome based on values of a set of predictor variables. It is similar to a linear regression model but is suited to models where the dependent variable is dichotomous. The aim of this research was to apply both neural networks, discriminant function analysis (a more traditional statistical approach under the general linear model) and logistic regression and compare their ability as statistical techniques to classify the different genders based nine sports psychological constructs to measure motivations to participate in masters sports. The sample consisted of 3687 male and 3488 female master’s athletes who participated in the 2009 World Masters Games and represented a volunteer/convenient sample in the study and a cross-sectional non-experimental research design. The Motivations of Marathoners Scales (MOMS) psychometric instrument assessed participant motivation by nine constructs/factors using factor scores from a 56 item seven Likert type survey instrument measuring motivations to participate. These factors were health orientation, weight concern, personal goal achievement, competition, recognition, affiliation, psychological coping, life meaning and self-esteem. The accuracy of the solutions were assessed with neural networks, by classification accuracy using both test and holdout samples, predicted-byobserved chart, ROC curve, cumulative gains and lift charts, independent variable importance and normalised importance; and discriminant function analysis by both original and cross-validation samples, lambda values, p-values, tolerance, F to remove and in stepwise discriminant analysis by the hierarchy of inclusion steps. Similar methods were applied when assessing classification accuracy using logistic regression. The results in terms of MLP analysis was overall correct percent of 64.4% and the order of importance was competition, selfesteem, affiliation, recognition, weight concern, health orientation, goal achievement, psychological coping and life meaning. In terms of RBF analysis training sample overall correct percent was 60.5% and order of importance was competition, affiliation, recognition, psychological coping, weight concern, life, meaning, self-esteem, goal achievement and health orientation. For the discriminant analysis the overall correct classification rate was 63.3% and for logistic regression 63.0% and the stepwise entry order into both analyses was affiliation, competition, self-esteem, recognition, weight concern and health orientation. The classification accuracies based on MLP, discriminant analysis and logistic regression were very similar in outcome for both the classification of gender and combined classification accuracy. None of the classification techniques based on neural network analyses and multivariate method of discriminant analysis and logistic regression were overtly superior to each other. Although it is important to note the RBF neural network displayed classification accuracy slightly lower than the other three methods.

Two horticultural crops, the ornamental liriopogon and the sweetpotato [Ipomoea batatas (L.) Lam.], were analyzed for morphological, quantitative and molecular marker variation using Amplified Fragment Length Polymorphism (AFLP) and various multivariate statistical techniques. Ornamental cultivars in genera Liriope and Ophiopogon were analyzed for relatedness using AFLP marker data and statistical clustering methods. Marker data did not substantiate the separation of these two genera. Greater than 95 % of the total genetic variability present was attributed to within group variation (P < 0.05). Trait-linked molecular markers were identified using Quantitative trait loci (QTL) analysis, logistic regression and discriminant analysis in the studies involving sweetpotato. The traits studied included dry matter content, virus disease resistance, root-knot nematode resistance, sugar content and -carotene content. Analysis of molecular variance found significant (P <0.001) differences between two phenotypic groups from unrelated genotypes for dry matter data. Using 14 markers selected through discriminant analysis the phenotypic grouping was validated with a zero error rate. Eighty-seven F1 sweetpotato genotypes from a cross of ‘Tanzania’ and ‘Wagabolige’ landraces were used to generate AFLP and random amplified polymorphic DNA (RAPD) marker profiles for this study. One AFLP marker linked to sweetpotato chlorotic stunt closterovirus resistance and one RAPD marker linked to sweetpotato feathery mottle virus resistance previously identified by traditional mapping strategies were selected plus new markers. Two diverse F1 populations of sweetpotato were used to identify and select markers suitable for identification of plants possessing a resistant reaction to southern root-knot nematode race 3 [Meloidogyne incognita (Kofoid and White) Chitwood]. Results for plant nematode resistance indicated a binomial distribution among the genotypes for population 1 and a normal distribution for population 2. A comparison of the power of discriminant analysis models for southern root-knot nematode resistance class prediction achieved 88% classification efficiencies. An F1 population of 73 clones consisting of parents and half-sibs was grouped into 2 phenotypic classes based on their sugar and ß-carotene content. Logistic regression and discriminant analysis selected meaningful markers that had significant associations with each of the traits. These results validated discriminant analysis and logistic regression as meaningful trait-linked marker selection methods.

For researchers, banks' profitability has been a key factor. Various internal and external factors may have an impact on the banks’ profitability. Internal factors are unique to each bank and have an impact on management decisions that affect bank profitability. External factors are those outside of the banks' control that can affect how well they perform. This study examines significant financial variables identifying profitability perspectives and two predictive models of banks’ profitability using the banks' data from the period of 2007-2016. Logistic regression and discriminant analyses were used and evaluated their predictive accuracy. Although, logistic regression and discriminant analysis are unusual for such studies as the dependent variable is continuous and these techniques are applicable when the dependent variables are categorical. So, in order to convert the dependent variable i.e. return on assets to binary variable, ’1’ is assigned for a positive return on assets and ‘0’ for a negative return on assets. The empirical results described that there is a statistical difference between the prediction results of logistic and discriminant analyses. The discriminant models identified that non-interest income, interest rate, operating efficiency, and bank size have a significant effect on banks’ profitability. On the other hand logistic regression showed there are two significant factors i.e. interest ratio and bank size. In the comparison of the models and the accuracy rate of discriminant analysis is higher than that of logistic regression

Bayesian-logit model for risk assessment in coronary artery bypass grafting

In this study, it was aimed to determine the variables affecting students' proficiency classification by using the data of 4th-grade students participating in the TIMSS 2019 application in the fields of mathematics and science. For this purpose, it was tried to provide evidence for classification validity with the variables of school belonging, bullying, home resources for learning, self-efficacy for computer use, disorderly behavior in Math lessons, like learning Math/Science, confident in Math/Science and instructional clarity in Math/Science lessons. The study was conducted with a correlational design. The sample of the study consisted of 3887 students in both lessons, which remained as a result of the missing data deletion and assignment processes from 4028 students who originally participated in the application. Logistic regression and discriminant analysis were used to analyze the data. As a result of the study, it was determined that 41.6% for Mathematics and 43% for Science in logistic regression analysis and 42.5% for Mathematics and 45% for Science in discriminant analysis were correctly classified through independent variables.

Summary1. Current emphasis on species conservation requires the development of specific distribution models. Several modelling methods are available, but their performance has seldom been compared. We therefore used discriminant analysis, logistic regression and artificial neural networks with environmental data to predict the presence or absence of six river birds along 180 Himalayan streams. We applied each method to calibration sites and independent test sites. With logistic regression, we compared performance in predicting presence–absence using map‐derived predictors (river slope and altitude) as opposed to detailed data from a standardized river habitat survey (RHS).2. Using the entire calibration data, overall success at predicting presence or absence was only slightly greater using artificial neural networks (89–100%) than either logistic regression (75–92%) or discriminant analysis (81–95%), and on this criterion all methods gave good performance.3. When applied to independent test data, overall prediction success averaged 71–80%, with logistic regression marginally but significantly out‐performing the other methods. Encouragingly for researchers with limited data, model performance in jack‐knife tests faithfully represented performance in more rigorous validations where calibration (n = 119) and test sites (n = 61) were in separate geographical regions.4. All three methods predicted true absences (83–92% success) better than true presences (31–44%). Results from logistic regression were the most variable across species, but positive prediction declined with increasing species rarity in each method.5. Applications with logistic regression illustrated that significant habitat predictors varied between data sets within species. Hypotheses about causal effects by habitat structure on distribution were thus difficult to erect or test. Logistic regression also showed that detailed data from the river habitat survey substantially improved positive prediction by comparison with prediction using slope or altitude alone.6. We conclude that discriminant analysis, logistic regression and artificial neural networks differ only marginally in performance when predicting species distributions. Model choice should therefore depend on the nature of the data, on the needs of any particular analysis, and on whether assumptions for each method are satisfied. All three methods share drawbacks due to systematic effects by species rarity on performance measures. They also share limitations due to the correlative nature of survey data often used for model development at the spatial scales required in macro‐ecology and conservation biology. Tests with independent data, using a wider range of performance measures than those used traditionally, will be important in examining models and testing hypotheses for such applications.

Chapter 32 - Machine Learning and Short Positions in Stock Trading Strategies

This study aims to evaluate the effects of natural and human conditions on vegetation covers using VCI and TCI indices (the satellite-based Vegetation Health Indices (VHIs)) in Dak Nong province. Factors affecting the VHIs included in the analysis involve land use, soil, population, topography, distance to roads and surface water. The data analysis period is the dry season from 2000 to 2020. The trend of the VHIs’ change in this study is analyzed using Sen’s method with monitoring data from Modis. The effects of factors affecting the VHIs are based on logistic regression and discriminant analysis. The analysis results show that the VHIs are clear and show both increasing and decreasing trends. Based on logistic regression analysis, the influence of land use types on the trend of the VHIs in the direction of increasing from negative to positive will be from PdF, UnL, AnC, SpF, PtF, PeA and then to PdR. Corresponding to the soil will be from ACa, FRp, FRx, FRr, ACh, LVx, FLg and then to LVg. Based on the discriminant analysis method, it was found that there are only four primary factors affecting the trend of the VHIs in order of decreasing level of closure: population density, land use, soil and population growth rate and to the road. These results show that in places where labour resources are available, plants are interested in investment due to high profits, nutrient-rich soil and convenience for plant care; the VHIs tend to increase and vice versa. Compared with VCI, the trend of TCI tends to be sloping negative and more pessimistic. Through the coefficients of the logistic regression equation and the difference function, the TCI is more sensitive than the VCI when the independent variables change, especially for changes in population density and land use. Thus, TCI can be considered the preferred option for assessing vegetation health trends in the context of climate change.

A population that consisted of F1 half-sibs and their parents was replicated in field plantings to generate roots to identify AFLP molecular markers and compare statistical models for selection of trait-linked markers using β-carotene content as the grouping variable. Genotypes were grouped into high and low β-carotene classes based on hierarchical cluster analysis. Association parameters between the high and low phenotypic classes of β-carotene and molecular marker profiles were made using logistic regression and discriminant analysis. Logistic regression selected 8 markers less that were associated with β-carotene content compared to discriminant analysis. Further analysis showed that logistic regression achieved 100% correct classification with the 9 markers selected compared to 17 markers that were required by discriminant analysis thus making logistic regression more efficient than discriminant analysis. Logistic regression based on Frequentist hypothesis testing differs from Bayesian based discriminant analysis. This difference may account for the percent correct classification observed.

Identification of the human body or remains after death is a forensic procedure which is difficult to perform and is mandatory by law and in compliance with social norms. Sexing the recovered human remains is an integral part of the identification process. The utilization of radiological imaging methods in anthropometric studies is being expanded by the application of modern imaging methods, leading to a decrease in costs, time, and the ability to create three-dimensional images. It is widely accepted that using population-specific criteria to estimate biological traits in the human skeleton improves accuracy. The present study investigated three-dimensional (3D) computed tomography images of 274 individuals (183 males and 91 females) within the 1884 years age group (Mean age: 43.13 ± 17.23) for sex determination. A total of 16 anthropometric parameters were taken into consideration. Logistic regression analysis (LRA), multiple logistic regression, univariate, multivariate, and stepwise discriminant function analysis (DFA) were generated to establish the accuracy of the cranial parameters. The morphometric measurements of the skull revealed that the most dimorphic structure was the bizygomatic breadth, which emerged as the best predictor of sex, with an accuracy of 80.7% with LRA and 81.2% with univariate DFA. Our study evaluates sexual dimorphism in an Indian population using CT skull images, incorporating cranial measurements, logistic regression, and univariate and multivariate analyses to develop an anthropometric approach for sex determination. The uniqueness of this research lies in its focus on the Indian population, where morphological variations may differ from other populations. By refining established CT-based sex determination methods and incorporating advanced quantification techniques, we aim to enhance the accuracy of sex determination in this population.

The use of discriminant analysis, logistic regression and classification tree analysis in the development of classification models for human health effects