Two-group Discriminant Analysis Research Articles

Exploring Psychological and Socio-Economic Impact of Covid 2019 on Healthcare Workers in India: Evidence from Delhi-NCR Hospitals

The prevalence and severity of Covid 2019 has brought to light new vulnerabilities of Health Care Workers (HCWs) in their working conditions, social protection, and occupational safety and mental health. In view of this, our paper aimed to provide comprehensive analysis of psychological stress, socio-economic impact, government intervention, interpersonal relations, work-life balance, and coping strategies of HCWs during the pandemic time in India. Based on data collected physically from 606 HCWs working in seven hospitals located in Delhi-NCR using structured questionnaires, six latent variables were constructed. T-test and ANOVA technique along with Post-hoc analysis was used to make comparisons and identify groups across which significant differences existed. Further, a two-group discriminant analysis was performed to find the variables that best discriminated between HCWs from private or public hospitals and medically or not-medically trained. Lastly, chi-squared (χ²) test was performed to examine the statistical association between coping strategies and few demographic characteristics. The results revealed that females and medically-trained HCWs were more psychologically stressed than their counterparts. Work-life-balance and interpersonal relations were found to be more disturbed among the medically-trained HCWs. Government regulations favourably affected the non-medically trained and HCWs with children and elderlies. Additionally, the impact of interpersonal relationships and work-life-balance was found to be adverse on permanent HCWs. Discriminant analysis showed that interpersonal relationships, followed by government intervention, work-life balance and socio-economic impact were the most important predictors between medically and non-medically trained HCWs. However, none of these factors discriminated between HCWs from private or public hospitals. Further, the preference patterns of coping strategies suggested that HCWs worked with an optimistic state of mind and adopted positive techniques. To make the healthcare system more resilient, it is strongly suggested that the government should play a key role during the times of pandemic, especially-encompassing females and financially stressed categories of HCWs.

An exploratory study to assess stakeholder's perception for new online teaching-learning environment during covid-19

Certain events bring unprecedented changes for the way things take place in our life. Spread of COVID-19 has brought a paradigm shift in a similar manner to the way we learn things and has dramatically altered the teaching learning process. Like any other tool, online learning was considered to be a supporting hand for augmenting the teaching learning experience by making the learning process more interactive and participative. COVID-19 has made online teaching as the sole mode for teaching learning for students and teachers. Without any doubt, when we run out of options only then we can ascertain the true facets of something. It can also be applied to evaluate the effectiveness of online teaching over traditional teaching. The present study is an endeavor to assess the stakeholder's perception for this new teaching learning environment which has emerged during this unprecedented situation. Respondents were selected using stratified judgment sampling method. 120 respondents from Delhi NCR were interviewed using structured questionnaire. Statistical tools like factor analysis, cross tabulation, t-test and discriminant analysis were used for data analysis. Factor analysis led to formation of dimensions namely awareness related to online teaching learning process, perception for decisions taken by higher education bodies, impact of pandemic on economic conditions and subsequent attendance and effectiveness of online classes. Further two-group discriminant analysis was applied to find out association among predictors and effectiveness of online learning. It was observed that awareness related to online teaching learning process, perception for decisions taken by administration, impact of pandemic on economic conditions, attendance, socio-economic background were prominent factors which significantly affect the respondent's perception for effectiveness of online classes. © 2021 Karadeniz Technical University. All rights reserved.

숙련성에 따른 스노보드 턴 동작의 제어전략

The purpose of this study is to understand the control strategic of the turn in snowboard according to expertise. Six experts and ten novices participated in this experiment. The task was to ride a snowboard, turn four cones, and down the slope. Four camcorders were used for recording. DLT method was used to get the 3D data. The data was analyzed with Kwon3D motion analysis system. Statistics was used to carry out the two-group discriminant analysis, and the significance level set at p<.05.<BR> results of this study were as follows: First, the novices showed an increasing rate of time performing as a characteristics of the task while experts showed a relatively constant rate performance for the phase of movement performance patterns. Second, the experts showed smaller angle chang of the shoulder line than novices in the phase 1. This indicates that the novice"s shoulder line movement is actively involved in the pattern of the degree of freedom released. On the other hand, in the phase 2, it can be seen that the experts is more involved in the task performance than the novices, as the degree of freedom in which angle chang of the knee line is larger than the novices. Third, the result of the coordination of the form of the alignment line for experts showed a strong coupled form of them and novices showed a released form. All subjects showed an anti-phase movement form.<BR> In conclusion, to control of snowboard movement, in phase two; the experts used a control strategy of lower body through the form of coupled coordination, which is significant then the involved within the novices. But, the novices used a control strategy to make turns with upper body movement in the phase 1 and showed a pattern of released coordination in all phases.

Fibromyalgia diagnostic model derived from combination of American College of Rheumatology 1990 and 2011 criteria.

BackgroundWe aimed to explore the American College of Rheumatology (ACR) 1990 and 2011 fibromyalgia (FM) classification criteria’s items and the components of Fibromyalgia Impact Questionnaire (FIQ) to identify features best discriminating FM features. Finally, we developed a combined FM diagnostic (C-FM) model using the FM’s key features.MethodsThe means and frequency on tender points (TPs), ACR 2011 components and FIQ items were calculated in the FM and non-FM (osteoarthritis [OA] and non-OA) patients. Then, two-step multiple logistic regression analysis was performed to order these variables according to their maximal statistical contribution in predicting group membership. Partial correlations assessed their unique contribution, and two-group discriminant analysis provided a classification table. Using receiver operator characteristic analyses, we determined the sensitivity and specificity of the final model.ResultsA total of 172 patients with FM, 75 with OA and 21 with periarthritis or regional pain syndromes were enrolled. Two steps multiple logistic regression analysis identified 8 key features of FM which accounted for 64.8% of variance associated with FM group membership: lateral epicondyle TP with variance percentages (36.9%), neck pain (14.5%), fatigue (4.7%), insomnia (3%), upper back pain (2.2%), shoulder pain (1.5%), gluteal TP (1.2%), and FIQ fatigue (0.9%). The C-FM model demonstrated a 91.4% correct classification rate, 91.9% for sensitivity and 91.7% for specificity.ConclusionsThe C-FM model can accurately detect FM patients among other pain disorders. Re-inclusion of TPs along with saving of FM main symptoms in the C-FM model is a unique feature of this model.

Consistency of test-based method for selection of variables in high-dimensional two-group discriminant analysis

This paper is concerned with selection of variables in two-group discriminant analysis with the same covariance matrix. We propose a test-based method (TM) drawing on the significance of each variable. Sufficient conditions for the test-based method to be consistent are provided when the dimension and the sample size are large. For the case that the dimension is larger than the sample size, a ridge-type method is proposed. Our results and tendencies therein are explored numerically through a Monte Carlo simulation. It is pointed that our selection method can be applied for high-dimensional data.

Isolated focal dystonia phenotypes are associated with distinct patterns of altered microstructure.

ObjectiveIsolated adult-onset focal dystonia is considered a network disorder with disturbances to the motor basal ganglia and cerebellar circuits playing a pathophysiological role, but why specific body regions become affected remains unknown. We aimed to use diffusion tensor imaging to determine if the two most common phenotypes of focal dystonia are associated with distinguishing microstructural changes affecting the motor network.MethodsFifteen blepharospasm patients, 20 cervical dystonia patients, and 30 age- and sex-matched healthy controls were recruited. Maps of fractional anisotropy and mean diffusivity were analyzed using a voxel-based approach and an automated region-of-interest technique to evaluate deep gray matter nuclei. Correlations between diffusion measures and dystonia severity were tested, and post hoc discriminant analyses were conducted.ResultsVoxel-based analyses revealed significantly reduced fractional anisotropy in the right cerebellum and increased mean diffusivity in the left caudate of cervical dystonia patients compared to controls, as well as lower fractional anisotropy in the right cerebellum in cervical dystonia patients relative to blepharospasm patients. In addition to reduced fractional anisotropy in the bilateral caudate nucleus of cervical dystonia patients relative to controls and blepharospasm patients, region-of-interest analyses revealed significantly reduced fractional anisotropy in the right globus pallidus internus and left red nucleus of blepharospasm patients compared to both controls and cervical dystonia patients. Diffusivity measures in the red nucleus of blepharospasm patients correlated with disease severity. In a three-group discriminant analysis, participants were correctly classified with only modest reliability (67–75%), but in a two-group discriminant analysis, patients could be distinguished from each other with high reliability (83–100%).ConclusionsDifferent focal dystonia phenotypes are associated with distinct patterns of altered microstructure within constituent regions of basal ganglia and cerebellar circuits.

Interval estimation in two-group discriminant analysis under heteroscedasticity for large dimension

ABSTRACTThis article is concerned with the problem of discriminating between two populations with heteroscedastic multivariate normal distributions based on an observation vector . We give the limiting distribution of the unbiased estimator for the log odds ratio of the posterior probabilities as the sample sizes Ni (i = 1, 2) and the dimension p go to infinity together with the ratio p/(Ni − 1) converging a finite non zero constant ci ∈ (0, 1) for the case in which the prior probabilities are equal. Approximated interval estimation for the log odds ratio is derived. Simulation results indicate that our estimation has good accuracy compared with the classical results.

PRM20 - Predictive Modeling to Assess Predictors of Treatment Success and Failure Among Combination Statin Therapy Patients

Combination statin therapy may help to further lower low-density lipoprotein cholesterol (LDL-C) better than monotherapy alone. The objective of this study was to apply predictive modeling methodology to determine the predictors of success and failure in achieving LDL-C goals after combination statin-fibrate therapy in patients diagnosed with hypertriglyceridemia (HTG). A large claims database was used to identify patients initiating a fibrate between January 2011 and December 2011 (index date). Diagnosis of HTG and the use of statins were confirmed within 6 months before the index date. A total of 622 patients were selected for the current analysis. Patients were categorized into very high risk, high risk, moderate risk, and low risk groups. Logistic regression and two-group discriminant analysis models based on 17 potential predictors for treatment success or failure were constructed. At index, the median triglyceride (TG) level among all patients was 95.5 mg/dL, LDL-C level was 92 mg/dL, and high-density lipoprotein (HDL) was 40 mg/dL. The mean age was 54 years. Two predictors were associated with combination statin-fibrate treatment success or failure and accounted for 5.3% of variance between groups. Low HDL (defined as <40 mg/dL) (OR=0.35; 95% CI, 0.20-0.59) and peripheral arterial disease (OR=0.10; 95% CI, 0.02-0.38) were significantly associated with treatment failure. Low HDL variable was the key discriminator. Analytic insights enabled by predictive models may help researchers gain information on discriminating factors about certain target treatment groups and drug classes. A set of key predictors may suggest opportunities to understand and predict treatment success and failure of targeted groups and/or drug classes. These predictors may be useful in developing treatment strategies that will optimize outcomes.

High-dimensional AICs for selection of variables in discriminant analysis

This paper is concerned with high-dimensional modifications of Akaike information criterion (AIC) for selection of variables in discriminant analysis. The AIC has been proposed as an asymptotically unbiased estimator of the risk function of the candidate model when the dimension is fixed and the sample size tends to infinity. On the other hand, Fujikoshi (2002) attempted to modify the AIC in two-group discriminant analysis when the dimension and the sample size tend to infinity. Such an estimator is called high-dimensional AIC, which is denoted by HAIC. However, its modification was obtained under a restrictive assumption, and furthermore, it was difficult to extend the method to multiple-group case. In this paper, by a new approach we propose HAIC which is an asymptotically unbiased estimator of the risk function in multiple-group discriminant analysis when both the dimension and the sample size tend to infinity, for a general class of candidate models. By simulation experiments it is shown that HAIC is more useful than other AIC type of criteria.

Validation of tumor markers in central nervous system germ cell tumors by real-time reverse transcriptase polymerase chain reaction using formalin-fixed paraffin-embedded tissues

The therapeutic protocols for treatment of germinomas and non-germinomatous germ cell tumors (NGGCTs) are completely different, so it is important to distinguish pure germinomas from NGGCTs. As it can be difficult to diagnose by morphology alone, immunohisto-chemistry (IHC) has been widely used as an ancillary test to improve diagnostic accuracy. However, IHC has limitations due to the misinterpretation of results or the aberrant loss of immunoreactivity. However, real-time RT-PCR has certain advantages over IHC, including its quantitative nature. The aim of our study was to evaluate the usefulness of real-time RT-PCR on formalin-fixed paraffin-embedded (FFPE) tissue blocks for the diagnosis of germ cell tumors of the central nervous system. We selected eight markers of germ cell tumors using a literature search, and validated them using real-time RT-PCR. Among them, POU5F1, NANOG and TGFB2 were statistically significant (P=0.05) in multiple comparisons (MANOVA) of three groups (pure germinomas, mature teratomas and malignant germ cell tumors). Two-group (pure germinomas and NGGCTs) discriminant analysis achieved a 70.0% success rate in cross-validation. We concluded that real-time RT-PCR using FFPE tissue has adequate validating power comparable to IHC in the diagnosis of central nervous system germ cell tumors; therefore, when IHC is not available, not conclusive or not informative, RT-PCR is a potential alternative to a repeat biopsy.

A model selection criterion for discriminant analysis of high-dimensional data with fewer observations

This paper is concerned with the problem of selecting variables in two-group discriminant analysis for high-dimensional data with fewer observations than the dimension. We consider a selection criterion based on approximately unbiased for AIC type of risk. When the dimension is large compared to the sample size, AIC type of risk cannot be defined. We propose AIC by replacing maximum likelihood estimator with ridge-type estimator. This idea follows Srivastava and Kubokawa (2008). It has been further extended by Yamamura et al. (2010). Simulation revealed that the proposed AIC performs well.

Preschool Children's Attitudes Toward Older Adults: Comparison of Intergenerational and Traditional Day Care

Increasingly, attention has focused on the impact of intergenerational programming. Using a static group comparison design, this study examines preschoolers' attitudes toward older adults by comparing children in a shared site intergenerational program and those in traditional day care. Children were shown sketches of an older man and an older woman and were asked to describe them in terms of specific attributes. A two-group discriminant analysis was conducted to assess differences between the two programs yielding significant differences, with 82% of children correctly classified regarding program participation. Overall, attitudes toward older adults were more positive in the intergenerational program compared to the traditional program. In addition, preschoolers in the intergenerational program viewed older adults as healthy compared with preschoolers in a traditional program. Implications of these findings are examined.

Identification of plasma and hepatic parameters related to metabolic robustness in dairy cows

Blood plasma and hepatic parameters were identified that describe the differences between metabolically robust or vulnerable dairy cows grouped according to their past health status. Data from a field study on dairy cows were used from which metabolically challenged dairy cows were selected that had a milk fat percentage of >4.5 mg/g and a fat to protein ratio of >1.5 in their previous early lactation. The selected cows were either classified as metabolically robust or vulnerable based on the occurrence of various metabolic and (re)production disorders in their previous lactations. Blood and liver tissue samples were collected in week 3 ante partum (a.p.) (-3 wk), in week 4 (+4 wk) and in week 13 (+13 wk) post-partum (p.p.). Plasma concentrations of metabolites and hormones and mRNA expression of genes involved in metabolic pathways in the liver were used as variables for a two-group discriminant analysis (DA). Average discriminant scores (centroids) were different (p < 0.05) in -3 wk, +4 wk and in +13 wk. In -3 wk, significant variables that best explained the differences between metabolically robust and vulnerable cows were parity, plasma triglycerides, glucose and mRNA abundance of carnitine palmitoyltransferase 2 (CPT2). In addition, based on the classification matrix, 69% of the dairy cows were correctly classified. In +4 wk, identified significant parameters were parity, plasma glucose and urea, and 67% of the cows were correctly classified. In +13 wk, significant variables that explained the differences between the groups were parity, mRNA abundance of acyl-CoA synthetase long-chain 1 and CPT1, and 66% of the cows were correctly classified. In conclusion, the identified variables may distinguish from metabolically challenged cows, those cows that had a poorer health performance in their previous lactations.

Distinguishing fibromyalgia from rheumatoid arthritis and systemic lupus in clinical questionnaires: an analysis of the revised Fibromyalgia Impact Questionnaire (FIQR) and its variant, the Symptom Impact Questionnaire (SIQR), along with pain locations

IntroductionThe purpose of this study was to explore a data set of patients with fibromyalgia (FM), rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) who completed the Revised Fibromyalgia Impact Questionnaire (FIQR) and its variant, the Symptom Impact Questionnaire (SIQR), for discriminating features that could be used to differentiate FM from RA and SLE in clinical surveys.MethodsThe frequency and means of comparing FM, RA and SLE patients on all pain sites and SIQR variables were calculated. Multiple regression analysis was then conducted to identify the significant pain sites and SIQR predictors of group membership. Thereafter stepwise multiple regression analysis was performed to identify the order of variables in predicting their maximal statistical contribution to group membership. Partial correlations assessed their unique contribution, and, last, two-group discriminant analysis provided a classification table.ResultsThe data set contained information on the SIQR and also pain locations in 202 FM, 31 RA and 20 SLE patients. As the SIQR and pain locations did not differ much between the RA and SLE patients, they were grouped together (RA/SLE) to provide a more robust analysis. The combination of eight SIQR items and seven pain sites correctly classified 99% of FM and 90% of RA/SLE patients in a two-group discriminant analysis. The largest reported SIQR differences (FM minus RA/SLE) were seen for the parameters "tenderness to touch," "difficulty cleaning floors" and "discomfort on sitting for 45 minutes." Combining the SIQR and pain locations in a stepwise multiple regression analysis revealed that the seven most important predictors of group membership were mid-lower back pain (29%; 79% vs. 16%), tenderness to touch (11.5%; 6.86 vs. 3.02), neck pain (6.8%; 91% vs. 39%), hand pain (5%; 64% vs. 77%), arm pain (3%; 69% vs. 18%), outer lower back pain (1.7%; 80% vs. 22%) and sitting for 45 minutes (1.4%; 5.56 vs. 1.49).ConclusionsA combination of two SIQR questions ("tenderness to touch" and "difficulty sitting for 45 minutes") plus pain in the lower back, neck, hands and arms may be useful in the construction of clinical questionnaires designed for patients with musculoskeletal pain. This combination provided the correct diagnosis in 97% of patients, with only 7 of 253 patients misclassified.

COMPARING OPTIMISM OF ERROR RATE ESTIMATORS IN DISCRIMINANT ANALYSIS BY MONTE CARLO SIMULATION ON MULTIVARIATE NORMAL DATA

The problem considered in this paper is estimation of the error rate in two-group discriminant analysis. Here, performance of 19 existing error rate estimators are compared and contrasted by mean of Monte Carlo simulations under the ideal condition that both parent populations are multivariate normal with common covariance matrix. The criterion used for comparing those error rate estimators is optimism. Five experimental factors are considered for the simulation, they are the number of variables, the sample size relative to the number of variables, the Mahalanobis squared distance between the two populations, dependency factor among variables, and the degree of variation among the elements of the mean vector of the populations. The result of the simulation shows that there is no estimator performing the best for all situations. However, in general, the estimator U¹ proposed by Lachenbruch and Mickey (1968) is the best

MONTE CARLO EVALUATION OF ERROR RATE ESTIMATORS IN DISCRIMINANT ANALYSIS UNDER MULTIVARIATE NORMAL DATA

This paper is concerned with the problem of estimating the error rate in two-group discriminant analysis. Here, behaviour of 19 existing error rate estimators are compared and contrasted by mean of Monte Carlo simulations under the ideal condition that both parent populations are multivariate normal with common covariance matrix. The criterion used for comparing those error rate estimators is sum squared error (SSE). Five experimental factors are considered for the simulation, they are the number of variables, the sample size relative to the number of variables, the Mahalanobis squared distance between the two populations, dependency factor among variables, and the degree of variation among the elements of the mean vector of the populations. The result of the simulation shows that there is no estimator performing the best for all situations. However, on overall, the Finite Mixture Balanced bootstrap estimator (FMB) proposed by Mangku (2007) is the best estimator.

Empirical comparison of error rate-estimators in logistic discriminant analysis

A simulation study related to the comparison of 15 estimators of the actual-error rate in two-group logistic discriminant analysis is achieved in 528 cases taking into account some characteristics of the groups such as the common distribution of the variables, the overlap of the populations, the number of variables, etc. For each combination of these parameters, the relative errors of the estimators with regard to the theoretical actual-error rate (empirically determined) are computed and help to compare the performance of the estimators. The results show an overall best performance of the estimators e632 and eOS. On the contrary, the estimators eA, epp and eppcv record the lowest performance in terms of the mean relative error and the mean relative bias. Moreover, the performance of most of the parametric estimators depends on the normality and heteroscedasticity degrees of the populations.

An additive utility mixed integer programming model for nonlinear discriminant analysis

Mathematical programming (MP) discriminant analysis models can be used to develop classification models for assigning observations of unknown class membership to one of a number of specified classes using values of a set of features associated with each observation. Since most MP discriminant analysis models generate linear discriminant functions, these MP models are generally used to develop linear classification models. Nonlinear classifiers may, however, have better classification performance than linear classifiers. In this paper, a mixed integer programming model is developed to generate nonlinear discriminant functions composed of monotone piecewise-linear marginal utility functions for each feature and the cut-off value for class membership. It is also shown that this model can be extended for feature selection. The performance of this new MP model for two-group discriminant analysis is compared with statistical discriminant analysis and other MP discriminant analysis models using a real problem and a number of simulated problem sets.

Sex diagnosis of subadult specimens from Medieval Polish archaeological sites: Metric analysis of deciduous dentition

The subject of this work is the characterisation of the metric features of deciduous dentition in a Medieval population of central Poland with the use of the jackknife technique leave one out (LOO)—supporting multivariate methods, which are important for deriving discrimination equations that would result in sex determination of children's skeletal remains. The sex of the individuals was assessed through analysis of sex-specific DNA sequences (AMELY/AMELX, SRY and alpha satellite sequences). Discriminant analysis concerned only teeth of those individuals whose sex was confirmed by the primary structure of three DNA sequences. The deciduous tooth diameters of males were found to be significantly larger than those of females in four respects: MD diameter of the maxillary second molar, MD and BL diameters of the mandibular first molar and BL diameter of the mandibular second molar. A two-group discriminant analysis considered all those measurements as independent variables. A multiple regression procedure produced a linear equation predicting the sex of children's skeletons with a significant probability amounting to approximately 78%. The accuracy of the sex assessment of an individual, using dental measurements, was established at 69% in deciduous male and 88% in deciduous female teeth.

Significant characteristics of the urban rail renaissance in the United States: A discriminant analysis

In the rebirth of light-rail in the US, there has been little quantitative work detailing the differences between cities that have built rail transit and those that have not. In this study, 18 independent variables measuring a variety of characteristics that might promote or hinder rail transit construction are examined for 13 cities that built rail and 22 that did not, but considered it. After isolating the most significant variables, a two-group discriminant analysis generates a function from a randomly chosen set of 25 cities, and then cross-validates it on a separate set of 10. That model attempts to classify the cities into their respective groups. A model with three significant independent variables is generated that correctly classifies 33 of the 35 cities. The results indicate that cities which chose to build rail already had relatively well-used bus systems. There also appears to be an image and economic development aspect associated with rail construction.