Multiple Quantile Regression Models Research Articles

Multiple yield curve modeling and forecasting using deep learning

AbstractThis manuscript introduces deep learning models that simultaneously describe the dynamics of several yield curves. We aim to learn the dependence structure among the different yield curves induced by the globalization of financial markets and exploit it to produce more accurate forecasts. By combining the self-attention mechanism and nonparametric quantile regression, our model generates both point and interval forecasts of future yields. The architecture is designed to avoid quantile crossing issues affecting multiple quantile regression models. Numerical experiments conducted on two different datasets confirm the effectiveness of our approach. Finally, we explore potential extensions and enhancements by incorporating deep ensemble methods and transfer learning mechanisms.

Normative ultrasound values for Achilles tendon thickness in the general population and patients with Achilles tendinopathy: A large international cross-sectional study.

The objective of the study was to obtain adjusted ultrasonographic reference values of the Achilles tendon thickness (maximum anterior-posterior distance) in adults without (previous) Achilles tendinopathy (AT) and to compare these reference values with AT patients. Six hundred participants were consecutively included, comprising 500 asymptomatic individuals and 100 patients with clinically diagnosed chronic AT. The maximum tendon thickness was assessed using Ultrasound Tissue Characterization. A multiple quantile regression model was developed, incorporating covariates (personal characteristics) that were found to have a significant impact on the maximum anterior-posterior distance of the Achilles tendon. A 95% reference interval (RI) was derived (50th, 2.5th-97.5th percentile). In asymptomatic participants median (95% RI) tendon thickness was 4.9 (3.8-6.9) mm for the midportion region and 3.7 (2.8-4.8) mm for the insertional region. Age, height, body mass index, and sex had a significant correlation with maximum tendon thickness. Median tendon thickness for the midportion region was calculated with the normative equation -2.1 + AGE × 0.021 + HEIGHT × 0.032+ BMI × 0.028 + SEX × 0.05. For the insertional region, the normative equation was -0.34 + AGE × 0.010+ HEIGHT × 0.018 + BMI × 0.022 + SEX × -0.05. In the equations, SEX is defined as 0 for males and 1 for females. Mean (95% CI) difference in tendon thickness compared to AT patients was 2.7 mm (2.3-3.2, p < 0.001) for the midportion and 1.4 mm (1.1-1.7, p < 0.001) for the insertional region. Compared to the asymptomatic population 73/100 (73%) AT patients exhibited increased tendon thickening, with values exceeding the 95% RI. This study presents novel reference values for the thickness of midportion and insertional region of the Achilles tendon, which were adjusted for personal characteristics. Our novel web-based openly accessible calculator for determining normative Achilles tendon thickness (www.achillestendontool.com) will be a useful resource in the diagnostic process. Trial registration number: This trial is registered in the Netherlands Trial Register (NL9010).

Which Demographic and Clinical Characteristics Can Better Predict the Length of Hospital Stay among Traumatic Patients? A Retrospective Single-Center, Registry-Based Study.

Lengthy hospitalization may lead to an increased hospital-acquired patient complication, including infections, as well as increased costs for both healthcare systems and patients. A few studies evaluated the impact of various clinical and demographic variables on patients' length of stay (LOS). Hence, in this study, we aimed to investigate the impact of various variables on traumatic patients' LOS. This is a retrospective single-center, registry-based study of traumatic patients admitted to Taleqani, a major trauma center in Kermanshah, Iran. A Minimal Dataset (MDS) was developed to retrieve traumatic data on demographic and clinical aspects. We used univariable and multiple quantile regression models to evaluate the association between independent variables, including ISS, GCS, and SBP, with LOS. LOS is practically defined as the time interval between hospital admission and discharge. The LOS durations have been presented as median (Q1 to Q3) hours. A p-value of <0.05 was considered statistically significant. A total of 2708 cases were included in this study, with 1989 (73.4%) of them being male. The median LOS was 87.00 (48.00 to 144.00) hours. When adjusted for systolic blood pressure (SBP), Glasgow Coma Scale (GCS), Injury Severity Score (ISS), and cause of injury, the two characteristics of spine/back and multiple trauma were significantly associated with the higher LOS, with 43 (20.5 to 65.48) and 24 (10.39 to 37.60) hours more than extremities (P < 0.001 and P = 0.005). Besides, the patients admitted due to road traffic injuries (RTI) were discharged 16 and 41 hours later than falling and cutting/stabbing (P = 0.008 and < 0.001, respectively). Moreover, the patients with ISS≥16 and 9≤ISS≤15 had a median of 51 (21 to 80) and 34 (22 to 45) LOS hours more, compared to 1≤ISS≤8, respectively (P < 0.001). The trauma cases experiencing SBP ≤ 90 mmhg on admission had a median of 41 (20 to 62) hours more hospitalization period than those with SBP> 90 mmhg (P < 0.001). At last, the patients with GCS of 9 to 12 and GCS of 3 to 8 were hospitalized for 39 and 266 hours more than GCS of 13 to 15 (P < 0.001). Determining independent determinants of prolonged LOS may lead to better identifying at-risk patients on admission. Trauma care providers should consider the following risk factors for increased LOS: higher ISS, Lower GCS, and SBP, multiple trauma or spine injury, and trauma resulting from falling or cutting/stabbing. As a result, the impact of extended LOS might be reduced by intervening in the related influencing factors.

0785 Sleep Habits and Sleep Hygiene Practices of South African Primary School-Aged Children: An Exploratory Study

Abstract Introduction Insufficient sleep adversely affects the physiological and psychological well-being of children and their families. Reports indicate that children's sleep health is declining; however, comparable data on the South African pediatric population is lacking. To this end, the current study aimed to explore the sleep quality and sleep hygiene practices of South African primary school-aged children. Methods The current study was exploratory and employed convenience sampling. Parents/legal guardians of South African children (6 – 11 years) attending primary school were recruited through social media platforms and asked to complete an online survey. The survey captured basic biographical information on their child's sleep habits (Children’s Sleep Health Questionnaire, CSHQ) and sleep hygiene practices (Children’s Sleep Hygiene Scale, CSHS). Survey responses were collected from May–August 2022. Results Of the 448 responses, 208 were included in the data analysis. Responses were received from eight South African provinces; the highest response was from the Western Cape (38.5%), followed by Gauteng (30.3%). The mean age of the study population was 8.0 ± 1.7 years (mean ± SD) and 51% were female. Of the children surveyed, 49.5% attended government schools. The mean parental report of sleep duration was 9.31 ± 0.45 hrs. The mean total sleep disturbance score based on the CSHQ was 47 ± 8.4, with 76% of children scoring greater than or equal to 41, indicating the presence of a possible sleep disturbance. The mean sleep hygiene index was 28.0 ± 3.4. Linear regression analysis revealed a significant correlation between total sleep disturbance and sleep hygiene scores (R2 = 0.32, p &amp;lt; 0.001); low sleep hygiene index scores were associated with a higher total sleep disturbance score. Multiple quantile regression models revealed that home-schooling and increased age were also significantly associated with sleep disturbance scores. Conclusion While the reported total sleep time for South African primary school-aged children was within the recommended range (9 – 11 hrs) for this age group, the overall presence of possible sleep disturbances was high, which could be attributed to poor sleep hygiene practices. Support (if any)

Anemia is independently associated with mortality in people living with human immunodeficiency virus/acquired immune deficiency syndrome: A propensity score matching-based retrospective cohort study in China.

Although previous studies have suggested that hemoglobin is related to the health status of people living with human immunodeficiency virus/acquired immune deficiency syndrome (HIV/AIDS) (PLWHA), the role of anemia in mortality remains unclear. This study aimed to comprehensively quantify the effect of anemia on the mortality risk of PLWHA. In this retrospective cohort study, we thoroughly estimated the effect of anemia on PLWHA mortality, using data collected from January 2005 to June 2022 in the Huzhou area, in 450 subjects extracted from the database of the China Disease Prevention and Control Information System and matched them using a propensity score matching approach to balance potential confounding bias. The potential exposure-response relationship between anemia, hemoglobin concentration, and the mortality of PLWHA was also carefully estimated. A series of subgroup analyses, including interaction analysis, was further conducted to validate the robustness of the effect of anemia on PLWHA death risk. Anemia was significantly associated with an elevated death risk in PLWHA, with an increase of 74% (adjusted hazard ratio [AHR]: 1.74; 95% confidence interval [CI]: 1.03-2.93; p = 0.038) in those with anemia after adjusting for potential confounders. PLWHA with moderate or severe anemia had a higher risk of death, with an 86% increase (AHR = 1.86; 95% CI: 1.01-3.42; p = 0.045). Meanwhile, the AHR tended to increase by 85% on average (AHR = 1.85, 95% CI: 1.37-2.50; p < 0.001) with a per standard deviation (SD) decrease in plasma hemoglobin. Consistent relationships between plasma hemoglobin and the risk of death were further observed in the results from multiple quantile regression models, restricted cubic spline regression models, and a series of subgroup analyses. Anemia is an independent risk factor for HIV/AIDS-related mortality. Our findings may provide new insights into the relevance of PLWHA administration to public health policy, which demonstrate that this low-cost and routinely measured marker (hemoglobin) can be a marker of poor prognosis even before the start of HAART.

Relationship between Body Composition and Cardiac Autonomic Regulation in a Large Population of Italian Olympic Athletes.

Athletic performance is determined by many factors, such as cardiorespiratory fitness, muscular strength and psychological features, which all interact simultaneously. The large Italian National Olympic Committee database of Olympic athletes offers a unique healthy population to verify the strength of the interplay among a number of major elements of training, including autonomic nervous system (ANS) modulation, biochemical indicators and body composition, in a system medicine approach. This observational, retrospective study involved 583 individuals. As part of the yearly precompetitive examination, cardiac autonomic (heart rate variability), psychological, physical (cycloergometer stress test), biochemical and body composition (BOD POD) evaluations were performed. In subsequent analysis, we first considered the relationship between body composition and single individual variables in a simple correlation matrix, including a multitude of variables; then, Exploratory Factor Analysis (EFA) restricted the information to six latent domains, each combining congruent information in relation to body composition. Finally, we employed a multiple quantile regression model to evaluate possible relationships between ANSIs (index capable of synthetizing ANS regulation) and the latent domains indicated by EFA reflecting body composition. We observed a clear relationship between ANS and body mass composition parameters, as indicated by both bivariate correlations and the quantile regression result of ANSIs versus the latent domain aggregating mainly body composition data expressed in % (p = 0.002). In conclusion, these results suggest that specific training may elicit parallel adaptation of ANS control and body composition. The analysis of Olympic athletes’ data allowed us to obtain a better understanding of the complex, multidimensional factors involved in determining sport performance. The latter appears to be determined by the simultaneous interaction not only of cardiorespiratory fitness, muscular strength and psychological features, but also of ANS cardiovascular modulation and body composition.

The independent association of source-specific transportation noise exposure, noise annoyance and noise sensitivity with health-related quality of life

Noise exposure is affecting health-related quality of life (HRQoL). There are many modelling approaches linking specific noise sources with single health-related outcomes. However, an integrated approach is missing taking into account measured levels as well as noise annoyance and sensitivity and assessing their independent association with HRQoL domains. Therefore, we investigated the predictive association of most common transportation noise sources (aircraft, railway and road traffic) as well as transportation noise annoyance and noise sensitivity with HRQoL using data from SAPALDIA (Swiss Cohort Study on Air Pollution and Lung and Heart Diseases in Adults).We assessed 2035 subjects, who participated in the second and third wave of SAPALDIA (3&4) and had complete information on exposure, outcome and covariates. At SAPALDIA3, we calculated annual means (Lden) of source-specific transportation noise exposure at the most exposed facade of participant’s dwelling floor height. Participants reported noise annoyance on the widely used 11-point ICBEN scale and answered to 10 questions assessing individual noise sensitivity. To assess the potentially predictive effect of these noise exposures, HRQoL was assessed about 8 years later (SAPALDIA4) using the SF-36. We performed predictive multiple quantile regression models to elucidate associations of noise parameters measured at SAPALDIA3 with median SF-36 scores at SAPALDIA4.Source-specific transportation noise exposures showed few yet not consistent associations with HRQoL scores. We observed statistically significant negative associations of transportation noise annoyance with HRQoL scores covering mental health components (adjusted difference in SF-36 mental health score between highest vs. lowest annoyance tertile: −2.54 (95%CI: −3.89; −1.20). Noise sensitivity showed strongest and most consistent associations with HRQoL scores covering both general and mental health components (adjusted difference in SF-36 scores between highest vs. lowest sensitivity tertile: Mental health −5.96 (−7.57; −4.36); general health −5.16 (−7.08; −3.24)).Within all noise parameters, we predominantly observed negative associations of noise sensitivity with HRQoL attaining a magnitude of potential clinical relevance. This implies that factors other than transportation noise exposure may be relevant for this exposure-outcome relation. Nonetheless, transportation noise annoyance showed relevant associations with mental health components, indicating a negative association of transportation noise with HRQoL.

Performance Evaluation of the Multiple Quantile Regression Model for Estimating Spatial Soil Moisture after Filtering Soil Moisture Outliers

The spatial distribution of soil moisture (SM) was estimated by a multiple quantile regression (MQR) model with Terra Moderate Resolution Imaging Spectroradiometer (MODIS) and filtered SM data from 2013 to 2015 in South Korea. For input data, observed precipitation and SM data were collected from the Korea Meteorological Administration and various institutions monitoring SM. To improve the work of a previous study, prior to the estimation of SM, outlier detection using the isolation forest (IF) algorithm was applied to the observed SM data. The original observed SM data resulted in IF_SM data following outlier detection. This study obtained an average data removal rate of 20.1% at 58 stations. For various reasons, such as instrumentation, environment, and random errors, the original observed SM data contained approximately 20% uncertain data. After outlier detection, this study performed a regression analysis by estimating land surface temperature quantiles. The soil characteristics were considered through reclassification into four soil types (clay, loam, silt, and sand), and the five-day antecedent precipitation was considered in order to estimate the regression coefficient of the MQR model. For all soil types, the coefficient of determination (R2) and root mean square error (RMSE) values ranged from 0.25 to 0.77 and 1.86% to 12.21%, respectively. The MQR results showed a much better performance than that of the multiple linear regression (MLR) results, which yielded R2 and RMSE values of 0.20 to 0.66 and 1.08% to 7.23%, respectively. As a further illustration of improvement, the box plots of the MQR SM were closer to those of the observed SM than those of the MLR SM. This result indicates that the cumulative distribution functions (CDF) of MQR SM matched the CDF of the observed SM. Thus, the MQR algorithm with outlier detection can overcome the limitations of the MLR algorithm by reducing both the bias and variance.

Socio-demographic and clinical determinants of self-care in adults with type 2 diabetes: a multicentre observational study.

To describe self-care as defined by the Middle Range Theory of Self-Care of Chronic Illness and to identify clinical and socio-demographic determinants in a T2DM population. A multicentre observational cross-sectional study was conducted involving 540 adults with a confirmed diagnosis of T2DM from six outpatient diabetes services in Italy. Socio-demographic and clinical data were collected from medical records. The Self-Care of Diabetes Inventory (SCODI) was used to measure self-care maintenance, monitoring, management, and confidence dimensions. For each separate scale, scores were standardized 0-100 with higher SCODI scores indicating better self-care; a score ≥ 70 is adequate. Multiple quantile regression models were performed to identify determinants of each self-care dimension. Self-care maintenance (median = 81.3) and self-care confidence (median = 79.5) were adequate in most of the subjects. Self-care monitoring was adequate in only half of the sample (median = 70.6). Self-care management was poor (median = 59.4). Lower self-care maintenance was associated with lower self-care confidence (p < 0.001). Lower self-care monitoring was associated with being male (p < 0.001), having lower self-care confidence (p < 001), and having diabetes for < 10years (p < 0.001). Lower self-care management was associated with being male (p = 0.002), being older (p = 0.005), having a low income (p = 0.030), being employed (p = 0.008), having missed diabetes education in the last year (p = 0.002), and lower self-care confidence (p < 0.0001). Lower self-care confidence was associated with having diabetes for < 10years (p = 0.008), and having at least one comorbid condition (p = 0.006). Determinants of self-care maintenance, monitoring, management and confidence include both clinical and socio-demographic variables. Modifiable determinants such as self-care confidence and diabetes self-care management education could be used to tailor interventions to improve diabetes self-care.

복합 분위수 회귀에 대한 붓스트랩 방법의 응용

선형 회귀모형에서 오차항들이 서로 독립이고 동일한 분포를 따른다고 가정할 경우, (회귀계수의 강건한 추정을 위하여) 모든 분위수 함수의 회귀계수가 동일한 값을 갖는다는 사실에 근거한 복합 분위수 회귀(composite quantile regression) 방법을 고려할 수 있다. 본 논문에서는 복합 분위수 회귀에서 사용되는 분위수의 개수를 선택하기 위해 붓스트랩 방법의 가능성을 검토하였다. 또한, 분위수 회귀와 복합 분위수 회귀의 성능을 비교하기 위해 붓스트랩 방법을 이용하여 신뢰구간을 구축하고, 이들의 포함확률과 평균길이를 비교하였다. 이러한 모의실험을 통하여 복합 분위수 회귀의 우월성과 통계적 추론에 있어서 붓스트랩 방법의 유용성을 확인하였다. Composite quantile regression model is considered for iid error case. Since the regression coefficients are the same across different quantiles, composite quantile regression can be used to combine the strength across multiple quantile regression models. For the composite quantile regression, bootstrap method is examined for statistical inference including the selection of the number of quantiles and confidence intervals for the regression coefficients. Feasibility of the bootstrap method is demonstrated through a simulation study.