Parsimonious Model Research Articles

Contemporary Predictors of Major Adverse Cardiovascular Events following Percutaneous Coronary Intervention: A Nationally Representative US Sample

Background: Patient outcomes after percutaneous coronary intervention (PCI) have improved over the last 30 years due to better techniques, therapies, and care processes. This study evaluated contemporary predictors of post-PCI major adverse cardiovascular events (MACE) and summarized risk in a parsimonious risk prediction model. Methods: The Cardiovascular Patient-Level Analytical Platform (CLiPPeR) is an observational dataset of baseline variables and longitudinal outcomes from the American College of Cardiology’s CathPCI Registry® and national claims data. Cox regression was used to evaluate 2–6 years of patient follow-up (mean: 2.56 years), ending in December 2017, after index PCI between 2012 and 2015 (N = 1,450,787), to examine clinical and procedural predictors of MACE (first myocardial infarction, stroke, repeat PCI, coronary artery bypass grafting, and mortality). Cox analyses of post-PCI MACE were landmarked 28 days after index PCI. Results: Overall, 12.4% (n = 179,849) experienced MACE. All variables predicted MACE, with cardiogenic shock, cardiac arrest, four diseased coronary vessels, and chronic kidney disease having hazard ratios (HRs) ≥ 1.50. Other major predictors of MACE were in-hospital stroke, three-vessel disease, anemia, heart failure, and STEMI presentation. The index revascularization and discharge prescription of aspirin, P2Y12 inhibitor, and lipid-lowering medication had HR ≤ 0.67. The primary Cox model had c-statistic c = 0.761 for MACE versus c = 0.701 for the parsimonious model and c = 0.752 for the parsimonious model plus treatment variables. Conclusions: In a nationally representative US sample of post-PCI patients, predictors of longitudinal MACE risk were identified, and a parsimonious model efficiently encapsulated them. These findings may aid in assessing care processes to further improve care post-PCI outcomes.

Factors influencing consumers’ Airbnb use intention: a meta-analytic analysis using the UTAUT2

PurposeThis study integrates previous research on the intention to use Airbnb to determine which antecedents provide a parsimonious explanation.Design/methodology/approachMeta-analyses based on 61 samples estimate how 8 antecedents are associated with the intention to use Airbnb. Subsequent analyses utilize meta-analyses to estimate a regression model to simultaneously estimate the relationship between the antecedents and the intention to use Airbnb. Relative weight analysis then determined each antecedent’s utility.FindingsA parsimonious model with only four antecedents (hedonic motivation, price value, effort expectancy and social influence) was nearly as predictive as the full eight-antecedent model. Ten moderating variables were examined, but none were deemed to consistently influence the relationships between the antecedents and the intention to use Airbnb.Practical implicationsRelatively few measures (i.e. four) effectively explain customers’ intentions to use Airbnb. When these measures cannot be readily influenced, alternatives are also presented. Implications for the travel industry are considered and straightforward approaches to increasing users are presented.Originality/valueThis is the first integrative review of customers’ intentions to use Airbnb. We integrate what is currently known about customers’ intentions to use Airbnb and then provide a robust model for Airbnb use intentions that both researchers and practitioners can utilize.

Conceptualizing Integration in Mixed Methods Research

Integration lies at the heart of mixed methods research, but it has primarily been considered from a practical rather than a conceptual or theoretical perspective. In this paper, I develop a contextually situated, parsimonious conceptual model of the substantive dimensions that constitute integration. I conclude that integration between heterogeneous components in mixed methods research is constituted by interdependence that is connected, transactional, transformative, coherent, and dynamic. This conceptual model is presented here in the expectation that it will prompt researchers to think more deeply about the phenomena they study and the integration of the methods they use to study those phenomena.

Inflammatory subphenotypes previously identified in ARDS are associated with mortality at intensive care unit discharge: a secondary analysis of a prospective observational study

BackgroundIntensive care unit (ICU)-survivors have an increased risk of mortality after discharge compared to the general population. On ICU admission subphenotypes based on the plasma biomarker levels of interleukin-8, protein C and bicarbonate have been identified in patients admitted with acute respiratory distress syndrome (ARDS) that are prognostic of outcome and predictive of treatment response. We hypothesized that if these inflammatory subphenotypes previously identified among ARDS patients are assigned at ICU discharge in a more general critically ill population, they are associated with short- and long-term outcome.MethodsA secondary analysis of a prospective observational cohort study conducted in two Dutch ICUs between 2011 and 2014 was performed. All patients discharged alive from the ICU were at ICU discharge adjudicated to the previously identified inflammatory subphenotypes applying a validated parsimonious model using variables measured median 10.6 h [IQR, 8.0–31.4] prior to ICU discharge. Subphenotype distribution at ICU discharge, clinical characteristics and outcomes were analyzed. As a sensitivity analysis, a latent class analysis (LCA) was executed for subphenotype identification based on plasma protein biomarkers at ICU discharge reflective of coagulation activation, endothelial cell activation and inflammation. Concordance between the subphenotyping strategies was studied.ResultsOf the 8332 patients included in the original cohort, 1483 ICU-survivors had plasma biomarkers available and could be assigned to the inflammatory subphenotypes. At ICU discharge 6% (n = 86) was assigned to the hyperinflammatory and 94% (n = 1397) to the hypoinflammatory subphenotype. Patients assigned to the hyperinflammatory subphenotype were discharged with signs of more severe organ dysfunction (SOFA scores 7 [IQR 5–9] vs. 4 [IQR 2–6], p < 0.001). Mortality was higher in patients assigned to the hyperinflammatory subphenotype (30-day mortality 21% vs. 11%, p = 0.005; one-year mortality 48% vs. 28%, p < 0.001). LCA deemed 2 subphenotypes most suitable. ICU-survivors from class 1 had significantly higher mortality compared to class 2. Patients belonging to the hyperinflammatory subphenotype were mainly in class 1.ConclusionsPatients assigned to the hyperinflammatory subphenotype at ICU discharge showed significantly stronger anomalies in coagulation activation, endothelial cell activation and inflammation pathways implicated in the pathogenesis of critical disease and increased mortality until one-year follow up.

78 Estimates of genetic parameters for weight, wool, and carcass traits of Rambouillet sheep

Abstract The Rambouillet sheep breed is the foundation to most western range sheep in the U.S. Through the National Sheep Improvement Program (NSIP), Rambouillet breeders seek to improve economically relevant traits. However, the parameters used in the NSIP genetic evaluation were last estimated two decades ago. Our aim was to re-estimate genetic parameters for body weight (BW), carcass, and wool traits. Data were provided by NSIP, which included BW at birth (BWT; n = 28,817), weaning [WWT; n = 5,633; age: 68 (SD 16) day], early post-weaning [PWTE; n = 23,303; 128 (SD 22) d], late post-weaning [PWTL; n = 7,689; 201 (SD 45) d] and as yearlings [YWT; n = 5,830; 375 (SD 33) d]. Yearling ultrasound scans of 12th rib backfat depth (YCF; n = 2,447) and eye muscle depth (YEMD; n = 2,457) were also available. Yearling wool traits were fiber curvature (YFC; n = 6,009), fiber diameter (YFD; n = 9,880), fiber diameter coefficient of variation (YFDCV; n = 8,803), greasy fleece weight (YGFW; n = 11,859) and staple length (YSL; n = 11,270). The pedigree consisted of 36,297 individuals. Animals with phenotypic records were progeny of 333 sires and 1,727 dams. A succession of univariate animal models was fitted. The most comprehensive model included direct and maternal additive genetic effects, their genetic covariance, and maternal permanent environmental and residual effects. Phenotypes were pre-adjusted for birth-rearing type category and, as covariates, dam age and age at recording. Contemporary group—a concatenation of flock, year, season, management group, sex, and, excluding BWT, recording date—was the only fixed effect fitted in the model. Model selection was based on the log-likelihood ratio test, and Akaike and Bayesian information criteria. The data structure, namely dams and their daughters and granddaughters contributing data, allowed reliable testing of the maternal effect terms. The direct-maternal genetic covariance did not explain variation for any trait (P &amp;gt; 0.18). The most parsimonious model for BW traits included direct and maternal additive genetic, and maternal permanent environmental, effects. For carcass and wool traits, only direct additive genetic effects were included in the final model. Such was unsurprising, as maternal effects often diminish as animals age. For the final models fitted (Table 1), direct heritabilities (± SE) for body weights ranged from 0.14 ± 0.01 (BWT) to 0.38 ± 0.04 (PWTL); the maternal heritabilities ranged from 0.04 ± 0.02 (PWTL; YWT) to 0.13 ± 0.01 (BWT). For carcass traits, direct heritabilities were 0.13 ± 0.04 (YCF) and 0.31 ± 0.05 (YEMD). For wool traits, they ranged from 0.37 ± 0.02 (YSL) to 0.58 ± 0.02 (YFD). These re-estimated parameter values are consistent with those used by NSIP, with the exception of PWTL (greater) and YSL (lower). With genomic data accumulating in Rambouillet, further refinements in parameter estimation are planned.

Currency Devaluation and Economic Growth in Nigeria

This study examined currency devaluation and economic growth in Nigeria. Specifically, the study economic growth, balance and external debt on currency devaluation in Nigeria and tracked the direction of causality between currency devaluation and real economic variables in Nigeria. The study adopted the qualitative and quantitative research design. Secondary time series data covering 1990 to 2020 were obtained in the study. Estimation methods used in the study’s analysis includes descriptive statistics, correlation analysis, ARDL Co-integration analysis, parsimonious error correction model, granger causality test and other post estimation tests. Findings from the study revealed that gross domestic product exert positive significant impact on currency devaluation captured with currency devaluation both in the short and long run; balance of trade exert positive significant impact on Nigeria currency devaluation both in the short run and insignificant positive impact in the long run; external debt exerts negative significant impact on Nigeria currency devaluation in the short run; and positive significant impact in the long run; and a unidirectional relationship exists between external debt and currency devaluation, balance of trade and currency devaluation, external debt and balance of trade respectively. Premised on these findings, the study advocated that government in collaboration with monetary agencies should seek sound measures to reduce import and encourage exports towards maintaining positive trade balance and improve economic growth; government should create and introduce economic reforms or macroeconomic policies targeted at causing friendly business environment, boost economic output and improving the performance of the economy and government and the Central Bank of Nigeria should ensure debts contracted are deployed towards encouraging investment in exportable goods and services.

A statistical framework for a new Kavya-Manoharan Bilal distribution using ranked set sampling and simple random sampling

In survival and stochastic lifespan modeling, numerous families of distributions are sometimes considered unnatural, unjustifiable theoretically, and occasionally superfluous. Here, a novel parsimonious survival model is developed using the Bilal distribution (BD) and the Kavya-Manoharan (KM) parsimonious transformation family. In addition to other analytical properties, the forms of probability density function (PDF) and behavior of the distributions' hazard rates are analyzed. The insights are theoretical as well as practical. Theoretically, we offer explicit equations for the single and product moments of order statistics from Kavya-Manoharan Bilal Distribution. Practically, maximum likelihood (ML) technique, which is based on simple random sampling (SRS) and ranked set sampling (RSS) sample schemes, is employed to estimate the parameters. Numerical simulations are used as the primary methodology to compare the various sampling techniques.

Physical performance strongly predicts all-cause mortality risk in a real-world population of older diabetic patients: machine learning approach for mortality risk stratification.

Prognostic risk stratification in older adults with type 2 diabetes (T2D) is important for guiding decisions concerning advance care planning. A retrospective longitudinal study was conducted in a real-world sample of older diabetic patients afferent to the outpatient facilities of the Diabetology Unit of the IRCCS INRCA Hospital of Ancona (Italy). A total of 1,001 T2D patients aged more than 70 years were consecutively evaluated by a multidimensional geriatric assessment, including physical performance evaluated using the Short Physical Performance Battery (SPPB). The mortality was assessed during a 5-year follow-up. We used the automatic machine-learning (AutoML) JADBio platform to identify parsimonious mathematical models for risk stratification. Of 977 subjects included in the T2D cohort, the mean age was 76.5 (SD: 4.5) years and 454 (46.5%) were men. The mean follow-up time was 53.3 (SD:15.8) months, and 209 (21.4%) patients died by the end of the follow-up. The JADBio AutoML final model included age, sex, SPPB, chronic kidney disease, myocardial ischemia, peripheral artery disease, neuropathy, and myocardial infarction. The bootstrap-corrected concordance index (c-index) for the final model was 0.726 (95% CI: 0.687-0.763) with SPPB ranked as the most important predictor. Based on the penalized Cox regression model, the risk of death per unit of time for a subject with an SPPB score lower than five points was 3.35 times that for a subject with a score higher than eight points (P-value <0.001). Assessment of physical performance needs to be implemented in clinical practice for risk stratification of T2D older patients.

Assessing complex PTSD and PTSD: validation of the German version of the International Trauma Interview (ITI)

ABSTRACT Background: With the introduction of the ICD-11 into clinical practice, the reliable distinction between Posttraumatic Stress Disorder (PTSD) and Complex Posttraumatic Stress Disorder (CPTSD) becomes paramount. The semi-structured clinician-administered International Trauma Interview (ITI) aims to close this gap in clinical and research settings. Objective: This study investigated the psychometric properties of the German version of the ITI among trauma-exposed clinical samples from Switzerland and Germany. Method: Participants were 143 civilian and 100 military participants, aged M = 40.3 years, of whom 53.5% were male. Indicators of reliability and validity (latent structure, internal reliability, inter-rater agreement, convergent and discriminant validity) were evaluated. Confirmatory factor analysis (CFA) and partial correlation analysis were conducted separately for civilian and military participants. Results: Prevalence of PTSD was 30% (civilian) and 33% (military) and prevalence of CPTSD was 53% (civilians) and 21% (military). Satisfactory internal consistency and inter-rater agreement were found. In the military sample, a parsimonious first-order six-factor model was preferred over a second-order two-factor CFA model of ITI PTSD and Disturbances in Self-Organization (DSO). Model fit was excellent among military participants but no solution was supported among civilian participants. Overall, convergent validity was supported by positive correlations of ITI PTSD and DSO with DSM-5 PTSD. Discriminant validity for PTSD symptoms was confirmed among civilians but low in the military sample. Conclusions: The German ITI has shown potential as a clinician-administered diagnostic tool for assessing ICD-11 PTSD and CPTSD in primary care. However, further exploration of its latent structure and discriminant validity are indicated.

Synesth: Comprehensive Syntenic Reconciliation with Unsampled Lineages

We present Synesth, the most comprehensive and flexible tool for tree reconciliation that allows for events on syntenies (i.e., on sets of multiple genes), including duplications, transfers, fissions, and transient events going through unsampled species. This model allows for building histories that explicate the inconsistencies between a synteny tree and its associated species tree. We examine the combinatorial properties of this extended reconciliation model and study various associated parsimony problems. First, the infinite set of explicatory histories is reduced to a finite but exponential set of Pareto-optimal histories (in terms of counts of each event type), then to a polynomial set of Pareto-optimal event count vectors, and this eventually ends with minimum event cost histories given an event cost function. An inductive characterization of the solution space using different algebras for each granularity leads to efficient dynamic programming algorithms, ultimately ending with an O(mn) time complexity algorithm for computing the cost of a minimum-cost history (m and n: number of nodes in the input synteny and species trees). This time complexity matches that of the fastest known algorithms for classical gene reconciliation with transfers. We show how Synesth can be applied to infer Pareto-optimal evolutionary scenarios for CRISPR-Cas systems in a set of bacterial genomes.

Using Empirical Information to Prioritize Early Literacy Assessment and Instruction in Preschool and Kindergarten

Early literacy skill development predicts later reading success, and development of skills in specific domains during the preschool years has been established as both a prerequisite and precursory for reading. Early literacy assessments typically include measures of separate skills across domains, and results can assist with determining where instructions may be most needed. When multiple areas of need are identified, understanding which skills to prioritize can be a challenge. Therefore, empirically identifying the relative contribution of each skill measured in preschool to subsequent reading success can promote more efficient systems of assessment. This study, conducted in the United States, examined the predictive validity of early literacy skills measured in preschool compared to skills measured in kindergarten, with a specific practical focus on identifying the most efficient predictive model for understanding reading readiness. Participants were 119 preschoolers (mean age = 66 months) who mostly spoke English as their primary language (79%). Results indicated early literacy and language skills in preschool are highly predictive of early reading in kindergarten, accounting for 59% of the variance in a reading composite score. The most parsimonious model indicated that first sounds, letter sounds, early comprehension, and expressive vocabulary measures adequately explained 52% of the variance in children’s kindergarten reading performance.

Case conceptualization in clinical practice and training

Case conceptualization is central to the success of the therapeutic process. However, integrative case conceptualization research has lagged behind research on integrating therapeutic intervention techniques. A successful case conceptualization provides (a) a dynamic, context-sensitive, yet parsimonious model of the client’s functioning; (b) relevant treatment targets and associated assessment procedures; and (c) a treatment plan including intervention phases and potential obstacles. Success in case conceptualization is a core clinical competency goal for trainees in clinical psychology and a career-long learning goal even for expert clinicians. Emerging technological trends and the formation of adversarial collaborative teams may assist research on the utility of well-constructed case conceptualizations.

Safety performance of dedicated and preferential bus lanes using multivariate negative binomial models for Bogotá, Colombia

Public transport priority systems such as Bus Rapid Transit (BRT) and Buses with High Level of Service (BHLS) are top-rated solutions to mobility in low-income and middle-income cities. There is scientific agreement that the safety performance level of these systems depends on their functional, operational, and infrastructure characteristics. However, there needs to be more evidence on how the different characteristics of bus corridors might influence safety. This paper aims to shed some light on this area by structuring a multivariate negative binomial model comparing crash risk on arterial roads, BRT, and BHLS corridors in Bogotá, Colombia. The analyzed infrastructure includes 712.1 km of arterial roads with standard bus service, 194.1 km of BRT network, and 135.6 km of BHLS network. The study considered crashes from 2015 to 2018 –fatalities, injuries, and property damage only– and 30 operational and infrastructure variables grouped into six classes –exposure, road design, infrastructure, public means of transport, and land use. A multicriteria process was applied for model selection, including the structure and predictive power based on [i] Akaike information criteria, [ii] K-fold cross-validation, and [iii] model parsimony. Relevant findings suggest that in terms of observed and expected accident rates and their relationship with the magnitude of exposure –logarithm of average annual traffic volumes at the peak hour (LOG_AAPHT) and the percentage of motorcycles, cars, buses, and trucks– the greatest risk of fatalities, injuries, and property damage occurs in the BHLS network. BRT network provides lower crash rates in less severe collisions while increasing injuries and fatalities. When comparing the BHLS network and the standard design of arterial roads, BHLS infrastructure, despite increasing mobility benefits, provides the lowest safety performance among the three analyzed networks. Individual factors of the study could also contribute to designing safer roads related to signalized intersection density and curvature. These findings support the unique characteristics and traffic dynamics present in the context of Bogotá that could inform and guide decisions of corresponding authorities in other highly dense urban areas from developing countries.

Multi-gated perimeter flow control for monocentric cities: Efficiency and equity

A control scheme for the multi-gated perimeter traffic flow control problem of monocentric cities is presented. The proposed scheme determines feasible and optimally distributed input flows for various gates located at the periphery of a protected network area. A parsimonious model is employed to describe the traffic dynamics of the protected network. To describe traffic dynamics outside of the protected area, the basic state-space model is augmented with additional state variables to account for vehicle queues at store-and-forward origin links at the periphery. The multi-gated perimeter flow control problem is formulated as a convex optimisation problem with finite horizon, and constrained control and state variables. This scheme aims to equalise the relative queues at origin links and to maintain the vehicle accumulation in the protected network around a desired set point, while the system’s throughput is maximised. For real-time control, the optimal control problem is embedded in a rolling-horizon scheme using the current state of the whole system as the initial state as well as predicted demand flows at origin/entrance links. Furthermore, practical flow allocation policies for single-region perimeter control strategies without explicitly considering entrance link dynamics are presented. These policies allocate a global perimeter-ordered flow to a number of candidate gates at the periphery of a protected network area by taking into account the different geometric characteristics of origin links. The proposed flow allocation policies are then benchmarked against the multi-gated perimeter flow control. A meticulous study is carried out for a 2.5 square mile protected network area of San Francisco, CA, including fifteen gates of different geometric characteristics. The results showed that the proposed approach is able to manage excessive queues outside of the protected network area and to optimally distribute the input flows, which confirms its efficiency and equity properties. Similar policies are expected to be utilised for dynamic routing and road pricing.

Trajectories of self-management and independence in youth with spina bifida: Family-related predictors of growth.

The purpose of this study was to assess family-related predictors of self-management trajectories in youth with spina bifida (SB). Participants with SB completed the Adolescent/Young Adult Self-Management and Independence Scale (AMIS II) interview across four time points. Family functioning, family-related stress, and perceived family support were assessed by multiple reporters and multiple methods. Growth in AMIS II total self-management and the AMIS II subscales (Condition and Independent Living) were estimated using linear mixed effect models as a function of family factors, after controlling for socio-demographic, condition-related, and neuropsychological variables that had been found to be significant predictors of self-management in prior studies. Model fit and parsimony were assessed using Akaike's information criterion (AIC). This diverse community sample included 99 respondents aged 18-27 years old. About half were female (52.5%) and White (52.5%); 15.2% were Black, and 32.3% were Hispanic/Latino. Observed family cohesion at baseline was associated with all self-management scales at age 18 (all p < 0.05). Growth in self-management was associated with parent-reported number of family stress events. For growth in total self-management, the best model included age, race/ethnicity, family income, shunt status, lesion level, neuropsychological function, observed family cohesion, and an age-by-number of family stress events interaction effect. The study findings suggested that family factors were important predictors of self-management trajectories, even after controlling for socio-demographic, condition-related, and neuropsychological covariates. Risk and protective factors identified in families of youth with SB can inform family-focused interventions for self-management.

On the approximation of high-order binary Markov chains by parsimonious models

Abstract We consider two parsimonious models of binary high-order Markov chains and discover their ability to approximate arbitrary high-order Markov chains. Two types of global measures for approximation accuracy are introduced, theoretical and experimental results are obtained for these measures and for the considered parsimonious models. New consistent statistical parameter estimator is constructed for parsimonious model based on two-layer artificial neural network.

Predicting weight gain in patients with cystic fibrosis on triple combination modulator.

Cystic fibrosis (CF) is caused by CF transmembrane conductance regulator (CFTR) gene mutations producing dysfunctional CFTR proteins leading to progressive clinical disease. Elexacaftor-tezacaftor-ivacaftor (ETI) remarkably improves lung disease but is associated with substantial weight gain. We performed a single-center longitudinal study predicting 6-month weight gain after ETI initiation. We used linear mixed effects modeling (LME) to determine association of ETI treatment with changing body mass index (BMI). Using linear regression, we examined BMI prediction models with distinct combinations of main effects to identify a model useful for patient counseling. We used up to eight commonly observed clinical characteristics as input variables (age, sex, percent predicted FEV1 [FEV1%], F508del homozygous state, pancreatic sufficiency, HgbA1c, prior modulator use and prior year number of pulmonary exacerbations). We evaluated 154 patients (19-73 years old, 54% female, FEV1% = 19-121, 0-6 prior year pulmonary exacerbations). LME demonstrated an association between ETI use and weight increases. Exhaustive testing suggested a parsimonious linear regression model well-fitted to data that is potentially useful for counseling. The two variable model shows that on average, BMI decreases by 0.045 (95% Confidence Interval [CI] = -0.069 to -0.021, p < 0.001) for every year of age and increases by 0.322 (CI = 0.142 to 0.502, p = 0.001) for each additional prior year exacerbation at the time of ETI initiation. Young patients with many prior year pulmonary exacerbations likely have the largest 6 month weight gain after starting ETI.

Nasopharyngeal carcinoma in children: a challenging disease in a middle-income country, Tunisia

Background: This study aimed to assess the epidemiological, clinical, and therapeutic aspects and prognosis of juvenile nasopharyngeal carcinoma in Tunis country.&#x0D; Materials and Methods: This study included 68 patients, younger than 18 years of age. All the patients had a clinical and para-clinical tumoral assessment. The study of survival and prognostic factors was done after a descriptive analysis. These prognostic factors were studied through uni and multivariate analysis.&#x0D; Results: The median age was 14.7 years and the sex ratio was 2 male to 1 female. The average time to first consultation was 4 months. Rhinological signs were the most frequent symptom for consultation (n= 41). The T3-T4 tumors accounted for 78% of patients and there was a lymph node invasion stage N2-N3 in 63% of cases. Non-metastatic patients had radiotherapy associated with chemotherapy in 97% of cases. Metastatic patients received hypofractionated radiotherapy on bone metastasis, and first-line chemotherapy followed by radiotherapy on the primitive tumor and lymph node areas in case of good response to chemotherapy ( n= 2). The mean follow-up was 94 months; 78% of these patients were alive and in complete remission, 19% were in therapeutic failure, and 16% of them had metachronous metastases. The five-year-overall survival was 95%. Hyposialia and skin dystrophy were the most frequent late complications.&#x0D; In univariate analyses, significant prognostic factors were cranial nerve invasion, intracranial invasion, and infra-temporal fossa invasion. In multivariate analysis, the most parsimonious model associated extension to the infratemporal fossa, endo-cranial extension, and initial therapeutic modality (treatment failures were less frequent with neoadjuvant chemotherapy (p= 0.22)). &#x0D; Conclusion: Treatment of nasopharyngeal carcinoma in children consists of chemotherapy and radiotherapy. Synchronous or metachronous metastases are common in this patient population. Modern radiotherapy techniques, including conformal radiotherapy with intensity modulation, are promising and could overcome toxicities in long-term survivors.

Locally time-varying parameter regression

. I discuss a framework to allow dynamic sparsity in time-varying parameter regression models. The conditional variances of the innovations of time-varying parameters are time varying and equal to zero adaptively via thresholding. The resulting model allows the dynamics of the time-varying parameters to mix over different frequencies of parameter changes in a data driven way and permits great flexibility while achieving model parsimony. A convenient strategy is discussed to infer if each coefficient is static or dynamic and, if dynamic, how frequent the parameter change is. An MCMC scheme is developed for model estimation. The performance of the proposed approach is illustrated in studies of both simulated and real economic data.

A framework for optimizing environmental covariates to support model interpretability in digital soil mapping

A common practice in digital soil mapping (DSM) is to incorporate many environmental covariates into a machine-learning algorithm to predict the spatial patterns of soil attributes. Variance inflation factor (VIF), principal component analysis (PCA), and recursive feature elimination (RFE) are three statistical methods that can be used to reduce the number of covariates. This study aims 1) to compare VIF and PCA approaches; 2) to identify an approach to determine the minimum number of covariates in DSM to ensure model parsimony using RFE after using VIF; and 3) to examine methods to interpret the impact of covariates on the variability of the predicted soil properties. The study area was the province of British Columbia (BC), Canada. This study used legacy data for four soil properties to make digital soil maps: soil organic carbon (SOC%), pH, clay%, and coarse fragment (CF%). Seven models were made for each soil property to determine the influence on validation results by using a different number of covariates produced by various methods on validation results. The results showed that the number of covariates could be reduced from 70 to 4 to 12 with only a little or no difference in concordance correlation coefficient (CCC) validation results. The CCC results of pH models using 70 and 7 covariates were both 0.74, and for other soil properties, this difference was negligible. The validation results obtained from PCA models showed that the performance of PCA in reducing the number of covariates was not as effective as when using VIF. Moreover, this study showed that covariates related to precipitation were the most important for modeling SOC%, soil pH, and clay%. Topographic covariates were the most influential covariates for modeling soil CF%. This study emphasizes the potential benefits of combining various data reduction methods to achieve optimal outcomes and generate the most parsimonious and interpretable models.