Categories
Nevin Manimala Statistics

GDP Forecasting: Machine Learning, Linear or Autoregression?

Front Artif Intell. 2021 Oct 15;4:757864. doi: 10.3389/frai.2021.757864. eCollection 2021.

ABSTRACT

This paper compares the predictive power of different models to forecast the real U.S. GDP. Using quarterly data from 1976 to 2020, we find that the machine learning K-Nearest Neighbour (KNN) model captures the self-predictive ability of the U.S. GDP and performs better than traditional time series analysis. We explore the inclusion of predictors such as the yield curve, its latent factors, and a set of macroeconomic variables in order to increase the level of forecasting accuracy. The predictions result to be improved only when considering long forecast horizons. The use of machine learning algorithm provides additional guidance for data-driven decision making.

PMID:34723174 | PMC:PMC8554645 | DOI:10.3389/frai.2021.757864

Categories
Nevin Manimala Statistics

Facing the Challenges of Developing Fair Risk Scoring Models

Front Artif Intell. 2021 Oct 14;4:681915. doi: 10.3389/frai.2021.681915. eCollection 2021.

ABSTRACT

Algorithmic scoring methods are widely used in the finance industry for several decades in order to prevent risk and to automate and optimize decisions. Regulatory requirements as given by the Basel Committee on Banking Supervision (BCBS) or the EU data protection regulations have led to an increasing interest and research activity on understanding black box machine learning models by means of explainable machine learning. Even though this is a step into a right direction, such methods are not able to guarantee for a fair scoring as machine learning models are not necessarily unbiased and may discriminate with respect to certain subpopulations such as a particular race, gender, or sexual orientation-even if the variable itself is not used for modeling. This is also true for white box methods like logistic regression. In this study, a framework is presented that allows analyzing and developing models with regard to fairness. The proposed methodology is based on techniques of causal inference and some of the methods can be linked to methods from explainable machine learning. A definition of counterfactual fairness is given together with an algorithm that results in a fair scoring model. The concepts are illustrated by means of a transparent simulation and a popular real-world example, the German Credit data using traditional scorecard models based on logistic regression and weight of evidence variable pre-transform. In contrast to previous studies in the field for our study, a corrected version of the data is presented and used. With the help of the simulation, the trade-off between fairness and predictive accuracy is analyzed. The results indicate that it is possible to remove unfairness without a strong performance decrease unless the correlation of the discriminative attributes on the other predictor variables in the model is not too strong. In addition, the challenge in explaining the resulting scoring model and the associated fairness implications to users is discussed.

PMID:34723172 | PMC:PMC8552888 | DOI:10.3389/frai.2021.681915

Categories
Nevin Manimala Statistics

Community SARS-CoV-2 seroprevalence before and after the second wave of SARS-CoV-2 infection in Harare, Zimbabwe

EClinicalMedicine. 2021 Nov;41:101172. doi: 10.1016/j.eclinm.2021.101172. Epub 2021 Oct 24.

ABSTRACT

BACKGROUND: By the end of July 2021 Zimbabwe, has reported over 100,000 SARS-CoV-2 infections. The true number of SARS-CoV-2 infections is likely to be much higher. We conducted a seroprevalence survey to estimate the prevalence of past SARS-CoV-2 in three high-density communities in Harare, Zimbabwe before and after the second wave of SARS-CoV-2.

METHODS: Between November 2020 and April 2021 we conducted a cross-sectional study of randomly selected households in three high-density communities (Budiriro, Highfield and Mbare) in Harare. Consenting participants answered a questionnaire and a dried blood spot sample was taken. Samples were tested for anti-SARS-CoV-2 nucleocapsid antibodies using the Roche e801 platform.

FINDINGS: A total of 2340 individuals participated in the study. SARS-CoV-2 antibody results were available for 70·1% (620/885) and 73·1% (1530/2093) of eligible participants in 2020 and 2021. The median age was 22 (IQR 10-37) years and 978 (45·5%) were men. SARS-CoV-2 seroprevalence was 19·0% (95% CI 15·1-23·5%) in 2020 and 53·0% (95% CI 49·6-56·4) in 2021. The prevalence ratio was 2·47 (95% CI 1·94-3·15) comparing 2020 with 2021 after adjusting for age, sex, and community. Almost half of all participants who tested positive reported no symptoms in the preceding six months.

INTERPRETATION: Following the second wave, one in two people had been infected with SARS-CoV-2 suggesting high levels of community transmission. Our results suggest that 184,800 (172,900-196,700) SARS-CoV-2 infections occurred in these three communities alone, greatly exceeding the reported number of cases for the whole city. Further seroprevalence surveys are needed to understand transmission during the current third wave despite high prevalence of past infections.

FUNDING: GCRF, Government of Canada, Wellcome Trust, Bavarian State Ministry of Sciences, Research, and the Arts.

PMID:34723165 | PMC:PMC8542175 | DOI:10.1016/j.eclinm.2021.101172

Categories
Nevin Manimala Statistics

Compact homogeneous Leviflat CR-manifolds

Complex Analysis Synerg. 2021;7(3-4):25. doi: 10.1007/s40627-021-00083-y. Epub 2021 Jul 15.

ABSTRACT

We consider compact Leviflat homogeneous Cauchy-Riemann (CR) manifolds. In this setting, the Levi-foliation exists and we show that all its leaves are homogeneous and biholomorphic. We analyze separately the structure of orbits in complex projective spaces and parallelizable homogeneous CR-manifolds in our context and then combine the projective and parallelizable cases. In codimensions one and two, we also give a classification.

PMID:34723129 | PMC:PMC8550170 | DOI:10.1007/s40627-021-00083-y

Categories
Nevin Manimala Statistics

Association between 25 Hydroxyvitamin D Concentrations and Lipid Profiles in Japanese with Type 2 Diabetes Mellitus

J Nutr Sci Vitaminol (Tokyo). 2021;67(5):266-272. doi: 10.3177/jnsv.67.266.

ABSTRACT

Low 25 hydroxyvitamin D (25(OH)D) levels are closely associated with the risk of cardiovascular disease. Vitamin D deficiency is more common in patients with type 2 diabetes mellitus than in the general population. In addition, vitamin D status is lower in patients with the metabolic syndrome than in those without the syndrome. Therefore, we examined the association between lipid profiles and 25(OH)D levels. In this case control study, 285 type 2 diabetic patients who attended the Manda Memorial Hospital from March to October 2017 were selected and 25(OH)D, total cholesterol, low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and triglyceride (TG) levels, were obtained. Multiple regression analysis revealed that the association between 25(OH)D concentrations and TG levels was statistically significant (p<0.01) after adjusting for age, sex, body mass index, estimated glomerular flow rate (eGFR), insulin use, duration of diabetes mellitus, glycosylated hemoglobin (HbA1c), alcohol consumption, current smoking, and sampling timing. The serum 25(OH)D level was inversely associated with the TG level after the adjustment for the characteristics of Japanese patients with type 2 diabetes mellitus.

PMID:34719611 | DOI:10.3177/jnsv.67.266

Categories
Nevin Manimala Statistics

Expectations and problems of the healthcare management support program for public assistance recipients

Nihon Koshu Eisei Zasshi. 2021 Oct 29. doi: 10.11236/jph.21-070. Online ahead of print.

ABSTRACT

Objectives In recent years, the importance of healthcare support for public assistance recipients has been recognized, and healthcare support measures have been implemented for them. This study aimed to investigate the expectations and problems of welfare offices, as well as their requests to the central government and prefectures about the healthcare management support program for public assistance recipients, which has been mandated since 2021.Methods In November 2019, snowball sampling was used to select 23 welfare offices for sending self-administered questionnaires about the healthcare management support program. Respondents were asked open-ended questions about their expectations and problems regarding the program, as well as their requests to the central government and prefectures. A subsequent interview survey was conducted from November 2019 to February 2020, gathering additional information on the questionnaire survey.Results We received consent for the questionnaire survey and interview survey from 16 welfare offices (response rate 69.6%). It was revealed that the staff in charge of the healthcare management support program at the welfare office expected the program to improve recipients’ health awareness and condition and for it to be applied to other residents in the community. They reported difficulty in developing the implementation system, setting up the indicators and target population, and retaining health professionals. They requested the central government and prefectures to clarify the indicators and the criteria for evaluation, provide reference materials, introduce precedents, communicate and coordinate with welfare offices and related organizations in the community, hold meetings to share information, and secure financial resources.Conclusion Findings from our study suggest a need to strengthen the cooperation between the central government, prefectures, and local governments and to establish a multilayered system to implement the healthcare management support program effectively in welfare offices.

PMID:34719539 | DOI:10.11236/jph.21-070

Categories
Nevin Manimala Statistics

Development of a risk prediction model and a sample risk chart for long-term care certification based on the functional health of older adults

Nihon Koshu Eisei Zasshi. 2021 Oct 29. doi: 10.11236/jph.21-022. Online ahead of print.

ABSTRACT

Objectives The first aim of this study was to develop risk prediction models based on age, sex, and functional health to estimate the absolute risk of the 3-year incidence of long-term care certification and to evaluate its performance. The second aim was to produce risk charts showing the probability of the incident long-term care certification as a tool for prompting older adults to engage in healthy behaviors.Methods This study’s data was obtained from older adults, aged ≥65 years, without any disability (i.e., they did not certify≥care level 1) and residing in Yabu, Hyogo Prefecture, Japan (n=5,964). A risk prediction model was developed using a logistic regression model that incorporated age and the Kihon Checklist (KCL) score or the Kaigo-Yobo Checklist (KYCL) score for each sex. The 3-year absolute risk of incidence of the long-term care certification (here defined as≥care level 1) was then calculated. We evaluated the model’s discrimination and calibration abilities using the area under the receiver operating characteristic curves (AUC) and the Hosmer-Lemeshow goodness-of-fit test, respectively. For internal validity, the mean AUC was calculated using a 5-fold cross-validation method.Results After excluding participants with missing KCL (n=4) or KYCL (n=1,516) data, we included 5,960 for the KCL analysis and 4,448 for the KYCL analysis. We identified incident long-term care certification for men and women during the follow-up period: 207 (8.2%) and 390 (11.3%) for KCL analysis and 128 (6.6%) and 256 (10.2%) for KYCL analysis, respectively. For calibration, the χ2 statistic for the risk prediction model using KCL and KYCL was: P=0.26 and P=0.44 in men and P=0.75 and P=0.20 in women, respectively. The AUC (mean AUC) in the KCL model was 0.86 (0.86) in men and 0.83 (0.83) in women. In the KYCL model, the AUC was 0.86 (0.85) in men and 0.85 (0.85) in women. The risk charts had six different colors, suggesting the predicted probability of incident long-term care certification.Conclusions The risk prediction model demonstrated good discrimination, calibration, and internal validity. The risk charts proposed in our study are easy to use and may help older adults in recognizing their disability risk. These charts may also support health promotion activities by facilitating the assessment and modification of the daily behaviors of older adults in community settings. Further studies with larger sample size and external validity verification are needed to promote the widespread use of risk charts.

PMID:34719536 | DOI:10.11236/jph.21-022

Categories
Nevin Manimala Statistics

The Educational Interpreter Supervision and Support Survey: Part 1

Am Ann Deaf. 2021;166(3):284-308. doi: 10.1353/aad.2021.0031.

ABSTRACT

Little literature exists documenting how K-12 American Sign Language-English interpreters are supervised or evaluated, or how supervision and evaluation affect interpreters and deaf and hard of hearing students. To address this gap, the author asked 230 educational interpreters about their experiences of supervision. Using a critical policy analysis framework and transformational leadership lens, the author conducted a Qualtrics online survey consisting of 42 open- and closed-ended items. Descriptive and statistical results are reported. Chi-square tests showed a significant difference in supervisory activities based on whether or not the supervisor was a credentialed interpreter. Supervisors who were interpreters conducted more job evaluations, assessed interpreting skills more often, and organized professional development more frequently than non-interpreters.

PMID:34719517 | DOI:10.1353/aad.2021.0031

Categories
Nevin Manimala Statistics

Deficient Interhemispheric Connectivity Underlies Movement Irregularities in Parkinson’s Disease

J Parkinsons Dis. 2021 Oct 23. doi: 10.3233/JPD-212840. Online ahead of print.

ABSTRACT

BACKGROUND: Movement execution is impaired in patients with Parkinson’s disease. Evolving neurodegeneration leads to altered connectivity between distinct regions of the brain and altered activity at interconnected areas. How connectivity alterations influence complex movements like drawing spirals in Parkinson’s disease patients remains largely unexplored.

OBJECTIVE: We investigated whether deteriorations in interregional connectivity relate to impaired execution of drawing.

METHODS: Twenty-nine Patients and 31 age-matched healthy control participants drew spirals with both hands on a digital graphics tablet, and the regularity of drawing execution was evaluated by sample entropy. We recorded resting-state fMRI and task-related EEG, and calculated the time-resolved partial directed coherence to estimate effective connectivity for both imaging modalities to determine the extent and directionality of interregional interactions.

RESULTS: Movement performance in Parkinson’s disease patients was characterized by increased sample entropy, corresponding to enhanced irregularities in task execution. Effective connectivity between the motor cortices of both hemispheres, derived from resting-state fMRI, was significantly reduced in PD patients in comparison to controls. The connectivity strength in the nondominant to dominant hemisphere direction in both modalities was inversely correlated with irregularities during drawing, but not with the clinical state.

CONCLUSION: Our findings suggest that interhemispheric connections are affected both at rest and during drawing movements by Parkinson’s disease. This provides novel evidence that disruptions of interhemispheric information exchange play a pivotal role for impairments of complex movement execution in Parkinson’s disease patients.

PMID:34719510 | DOI:10.3233/JPD-212840

Categories
Nevin Manimala Statistics

Knee Strength and Ankle Range of Motion Impacts on Timed Function Tests in Duchenne Muscular Dystrophy: In the Era of Glucocorticoids

J Neuromuscul Dis. 2021 Oct 29. doi: 10.3233/JND-210724. Online ahead of print.

ABSTRACT

BACKGROUND: Duchenne Muscular Dystrophy (DMD) is a neuromuscular disorder that presents in childhood and is characterized by slowly progressive proximal weakness and lower extremity contractures that limit ambulatory ability [1, 2]. Contractures develop in the ankles, knees, and hips due to muscle imbalances, fibrotic changes, loss of strength, and static positioning [2, 5]. Currently, standards of care guidelines emphasize the importance of maintaining good musculoskeletal alignment through stretching, bracing, and glucocorticoid (GC) therapy to preserve strength and function.

METHODS: This is a retrospective analysis of prospectively collected data through the CINRG Duchenne Natural history study (DNHS). The objectives of this analysis are to understand the progression of ankle contractures for individuals with DMD and to investigate the relationship between progressive lower limb contractures, knee strength, and Timed Function Tests.A collection of TFTs including supine to stand (STS), 10 meter walk test (10MWT), and timed stair climbing (4SC) have been used to monitor disease progression and are predictive of loss of ambulation in these patients [4]. Multiple factors contribute to loss of ambulation, including progressive loss of strength and contracture development that leads to changing biomechanical demands for ambulation. A better understanding of the changes in strength and range of motion (ROM) that contribute to loss of function is important in a more individualized rehabilitation management plan. In this longitudinal study, we measured strength using quantitative muscle testing (QMT) with the CINRG Quantitative Measurement System (CQMS)), ROM was measuresed with a goniometer and TFTs were measured using a standard stopwatch and methodology.

RESULTS: We enrolled 440 participants; mean baseline age was 8.9 (2.1, 28.0) years with 1321 observations used for analysis. GC use was stratified based on duration on drug with 18.7%at < 6 months or naïve; 4.3%<1 year; 58.0%1 < 10 years; and 19.3%between 10-25 years of GC use. Ankle ROM was better for those on GC compared to GC naive but did not significantly influence long-term progression rates. QMT, ROM, age and GCs contribute to speed of TFTs. Knee extension (KE) strength and Dorsiflexion (DF) ROM are significant predictors of speed for all TFTs (p < 0.001). Of the variables used in this analysis, KE strength is the primary predictor of walking speed, estimating that every pound increase in KE results in a 0.042 m/s improvement in 10MWT, and a smaller similar increase of 0.009 m/s with every degree of ankle DF ROM.

CONCLUSION: GC use provides an improvement in strength and ROM but does not affect rate of change. Knee strength has a greater influence on speed of TFTs than DF ROM, although both are statistically significant predictors of speed. Results show that retaining knee strength [1, 2], along with joint flexibility, may be important factors in the ability to perform walking, climbing and supine to stand activities.

PMID:34719507 | DOI:10.3233/JND-210724