Search

This paper provides two procedures to obtain prediction intervals for electricity demand and price based on functional data. The proposed procedures are related to one day ahead pointwise forecast. In particular, the first method uses a nonparametric autoregressive model and the second one uses a partial linear semi-parametric model, in which exogenous scalar covariates are incorporated in a linear way. In both cases, the proposed procedures for the construction of the prediction intervals use residual-based bootstrap algorithms, which allows also to obtain estimates of the prediction density. Applications to the Spanish Electricity Market, in year 2012, are reported. This work extends and complements the results of Aneiros et al. (2016), focused on pointwise forecasts of next-day electricity demand and price daily curves.

This paper presents an application of functional additive models in the context of electricity demand and price prediction. Data from the Spanish electricity market are used to obtain the pointwise predictions. Also prediction intervals, based on a bootstrap procedure, are computed. This approach is compared with the use of other functional regression methods applied to the same data set by Aneiros et al. (2016).

This paper deals with the semi-functional partial linear regression model $$Y={{\varvec{X}}}^\mathrm{T}{\varvec{\beta }}+m({\varvec{\chi }})+\varepsilon $$ under $$\alpha $$ -mixing conditions. $${\varvec{\beta }} \in \mathbb {R}^{p}$$ and $$m(\cdot )$$ denote an unknown vector and an unknown smooth real-valued operator, respectively. The covariates $${{\varvec{X}}}$$ and $${\varvec{\chi }}$$ are valued in $$\mathbb {R}^{p}$$ and some infinite-dimensional space, respectively, and the random error $$\varepsilon $$ verifies $$\mathbb {E}(\varepsilon |{{\varvec{X}}},{\varvec{\chi }})=0$$ . Naive and wild bootstrap procedures are proposed to approximate the distribution of kernel-based estimators of $${\varvec{\beta }}$$ and $$m(\chi )$$ , and their asymptotic validities are obtained. A simulation study shows the behavior (on finite sample sizes) of the proposed bootstrap methodology when applied to construct confidence intervals, while an application to real data concerning electricity market illustrates its usefulness in practice.

This study proposes a control chart based on functional data to detect anomalies and estimate the normal output of industrial processes and services such as those related to the energy efficiency domain. Companies providing statistical consultancy services in the fields of energy efficiency, heating, ventilation and air conditioning (HVAC), installation and control, and big data for buildings, have been striving to solve the problem of automatic anomaly detection in buildings controlled by sensors. Given the functional nature of the critical to quality (CTQ) variables, this study proposed a new functional data analysis (FDA) control chart method based on the concept of data depth. Specifically, it developed a control methodology, including the Phase I and II control charts. It is based on the calculation of the depth of functional data, the identification of outliers by smooth bootstrap resampling and the customization of nonparametric rank control charts. A comprehensive simulation study, comprising scenarios defined with different degrees of dependence between curves, was conducted to evaluate the control procedure. The proposed statistical process control procedure was also applied to detect energy efficiency anomalies in the stores of a textile company in the Panama City. In this case, energy consumption has been defined as the CTQ variable of the HVAC system. Briefly, the proposed methodology, which combines FDA and multivariate techniques, adapts the concept of the control chart based on a specific case of functional data and thereby presents a novel alternative for controlling facilities in which the data are obtained by continuous monitoring, as is the case with a great deal of process in the framework of Industry 4.0.

Background Obesity and atrial fibrillation ( AF ) frequently coexist and independently increase mortality. We sought to assess the association between obesity and adverse events in patients receiving oral anticoagulants for AF . Methods and Results Consecutive AF outpatients receiving anticoagulant agents (both vitamin K antagonists and direct oral anticoagulants) were recruited into the FANTASIIA (Atrial fibrillation: influence of the level and type of anticoagulation on the incidence of ischemic and hemorrhagic stroke) registry. This observational, multicenter, and prospective registry of AF patients analyzes the quality of anticoagulation, incidence of events, and differences between oral anticoagulant therapies. We analyzed baseline patient characteristics according to body mass index, normal: 2 , overweight: 25–30 kg/m 2 , and obese: ≥30 kg/m 2 ), assessing all‐cause mortality, stroke, major bleeding and major adverse cardiovascular events (a composite of ischemic stroke, myocardial infarction, and total mortality) at 3 years’ follow‐up. In this secondary prespecified substudy, the association of weight on prognosis was evaluated. We recruited 1956 patients (56% men, mean age 73.8±9.4 years): 358 (18.3%) had normal body mass index, 871 (44.5%) were overweight, and 727 (37.2%) were obese. Obese patients were younger ( P P =0.42). After a median follow‐up of 1070 days, 255 patients died (13%), 45 had a stroke (2.3%), 146 a major bleeding episode (7.5%) and 168 a major adverse cardiovascular event (8.6%). Event rates were similar between groups for total mortality ( P =0.29), stroke ( P =0.90), major bleeding ( P =0.31), and major adverse cardiovascular events ( P =0.24). On multivariate Cox analysis, body mass index was not independently associated with all‐cause mortality, cardiovascular mortality, stroke, major bleeding, or major adverse cardiovascular events. Conclusions In this prospective cohort of patients anticoagulated for AF , obesity was highly prevalent and was associated with more comorbidities, but not with poor prognosis.

The features of the electricity, as well as the rules of the competitive electricity markets, create the need of accurate predictions of electricity demand and price in order to anticipate decisions. Mainly, the prediction task in energy markets has been studied in the literature with the aim of obtaining scalar (hourly, daily, …) forecasts given scalar (and/or, less frequently, functional) historical data. This paper provides two methods to predict next-day electricity demand and price daily curves given information from past curves. They are based on using robust functional principal component analysis and nonparametric models with functional both response and covariate. In addition, the nonparametric proposal is extended to incorporate, in a linear way, exogenous scalar covariates. Results of these methods for the electricity market of mainland Spain, in year 2012, are reported. Their accuracy is compared with that of a naive method as well as with the corresponding to combining forecasts. Scalar versions are also included in the comparative study. This work extends and complements the methods and results in Vilar et al. (2012), focused on scalar forecasts.

Background One-third of atrial fibrillation (AF) patients have chronic kidney disease (CKD), a condition that itself increases thromboembolic and major bleeding risks, especially in patients with severe CKD. Bleeding would be accentuated by suboptimal anticoagulation control with vitamin K antagonists (VKA). Purpose This article aimed to investigate the incidence of cardiovascular events, mortality and quality of anticoagulation in relation to CKD in a ‘real-world’ prospective cohort of AF patients included in the FANTASIIA registry. Methods We analysed consecutive AF patients who were prospectively recruited with a year of follow-up. The quality of anticoagulation was estimated by time in therapeutic range (TTR). The annual incidence of events was analysed. Results We studied 1,936 patients (male: 55.7%, mean: 73.8 ± 9.4 years): 445 (22.9%) had normal function, 698 (36.1%) had mild CKD, 713 (36.8%) had moderate CKD and 80 (4.2%) had severe CKD. Patients with estimated glomerular filtration rate (eGFR) Conclusion In a prospective and real-world AF registry, approximately 67% of patients with severe CKD had poor anticoagulation control while taking VKA. The presence of severe CKD was an independent factor for cardiovascular mortality, MACE and major bleeding. Worsening eGFR of only ≥10 mL/min during follow-up was significantly associated with mortality and major bleeding.

Aims To describe the prevalence and associated factors of inappropriate doses of direct oral anticoagulants (DOAC) in a national registry of patients of real clinical practice. Methods and results Five hundred and thirty outpatients with atrial fibrillation treated with DOAC were included in a prospective, national, multicentre study. The appropriateness of the doses of DOAC was defined according to the recommendations of the European Heart Rhythm Association. Mean age was 73 ± 9 years, with a 46% of women. Two hundred and sixty-seven patients were prescribed dabigatran, 190 rivaroxaban, and 73 apixaban. A total of 172 patients (32%) did not receive the appropriate dose: 93 patients received a lower dose (18%) and 79 patients a higher dose (15%). In the comparisons among the subgroups of inappropriately low, appropriate, and inappropriately high dose, we observed significant trends to older age (69 ± 8 years vs. 73 ± 10 years vs. 77 ± 6 years), more frequent female sex (37% vs. 46% vs. 59%), antiplatelet drugs (5% vs. 8% vs. 25%), rivaroxaban (14% vs. 38% vs. 53%), and apixaban use (5% vs. 15% vs. 19%), higher CHAD2DS2-VASc (3.00 ± 1.38 vs. 3.58 ± 1.67 vs. 4.59 ± 1.44) and HAS-BLED scores (1.83 ± 0.87 vs. 1.92 ± 1.07 vs. 2.47 ± 1.13), lower body mass index (30 ± 6 kg/m2 vs. 29 ± 4 kg/m2 vs. 28 ± 4 kg/m2) and glomerular filtration rate (74 ± 27 mL/min vs. 70 ± 22 mL/min vs. 63 ± 16 mL/min), and lower frequency of dabigatran use (81% vs. 47% vs. 28%) (all comparisons P ≤ 0.01). Conclusion In this real-life study, 32% of patients received an inappropriate dose of DOAC. Several clinical factors can identify patients at risk of this situation. Sin financiación 5.047 JCR (2018) Q1, 30/136 Cardiac & Cardiovascular Systems 2.612 SJR (2018) Q1, 27/365 Cardiology and Cardiovascular Medicine, 7/108 Physiology (medical) No data IDR 2018 UEM

A functional depth measures the “centrality” of a functional datum (a function observed over a continuum, for example, a curve, an image) with respect to a given functional dataset. This paper proposes a way to detect outliers in functional time series based on functional depth. Ideally, the depth of a functional outlier should be very low but, when is it low enough to correspond to an outlier? This paper aims to address this question. It uses bootstrap techniques, which take into account the dependence between functional data, to solve this discriminant problem. Several Monte Carlo experiments were designed to explore the performance of the proposed procedure and compare it with some existing methods in the statistical literature related to independent functional data. The proposed methodology was finally used to detect outliers in temperature data and NOx emission data. Results conclude that dependence of data must be taken into account to detect outliers in functional time series. Copyright © 2015 John Wiley & Sons, Ltd.

The efficacy and safety of oral anticoagulation (OAC) using the vitamin K antagonists (VKA) are closely associated with the quality of anticoagulation, reflected by time in therapeutic range (TTR). The SAMe-TT2R2 is a risk score developed to predict the quality of anticoagulation control among VKA users. To analyse the quality of anticoagulation and its clinical determinants based on different methods in a prospective cohort of atrial fibrillation patients on VKA treatment participating in the multicentre Spanish observational registry FANTASIIA. Estimated TTR was calculated from Rosendaal, direct method, international normalized ratio variability, and NICE criteria. Time in therapeutic range values were compared for those patients with a SAMe-TT2R2 score 0-2 and >2. One thousand four hundred and seventy patients were analysed (56.4% male, mean age 74.1 ± 9.5 years). Mean TTR was 61.5 ± 25.1 with Rosendaal and 64.7 ± 24.2 with direct method. There was a high correlation between both methods (ρ = 0.805). The prevalence of poor anticoagulation control was 55%. Diabetes mellitus [odds ratio (OR) 1.38; P = 0.008], peripheral artery disease (PAD, OR 1.62; P = 0.048), and HAS-BLED (OR 1.13; P = 0.022) were independently associated with TTR 2. One thousand four hundred and seventy patients were analysed (56.4% male, mean age 74.1 ± 9.5 years). Mean TTR was 61.5 ± 25.1 with Rosendaal and 64.7 ± 24.2 with direct method. There was a high correlation between both methods (ρ = 0.805). The prevalence of poor anticoagulation control was 55%. Diabetes mellitus [odds ratio (OR) 1.38; P = 0.008], peripheral artery disease (PAD, OR 1.62; P = 0.048), and HAS-BLED (OR 1.13; P = 0.022) were independently associated with TTR 2 (P = 0.044), with a specificity of > 90% for predicting TTR 90% for predicting TTR In a multicentre prospective registry, 55% of AF patients had poor anticoagulation control with diabetes mellitus, PAD, and HAS-BLED being independently associated with TTR

Semi-functional partial linear regression model allows to deal with a nonparametric and a linear component within the functional regression. Naive and wild bootstrap procedures are proposed to approximate the distribution of the estimators for each component in the model, and their asymptotic validities are obtained in the context of dependence data, under α-mixing conditions. Based on that bootstrap procedures, confidence intervals can be obtained for each component in the model, which can be also extended to deal with prediction intervals and prediction densities.

[Abstract] We investigated the incidence of nonembolic adverse events in 2 cohorts of patients with atrial fibrillation (AF) and validated the 2MACE score ([metabolic syndrome, age ≥75] [doubled]; [myocardial infarction or revascularization, congestive heart failure {HF}, and stroke, transient ischemic attack or thromboembolism]) as predictor of major adverse cardiovascular events (MACEs). We recruited 2,630 patients with AF from 2 different cohorts (Murcia AF and FANTASIIA). The 2MACE score was calculated, and during a median of 7.2 years (Murcia AF cohort) and 1.01 years (FANTASIIA) of follow-up, we recorded all nonembolic adverse events and MACEs (composite of nonfatal myocardial infarction or revascularization and cardiovascular death). Receiver operating characteristic curves comparison, reclassification and discriminatory analyses, and decision curve analyses were performed to compare predictive ability and clinical usefulness of the 2MACE score against CHA2DS2-VASc. During follow-up, there were 65 MACEs in the Murcia cohort and 60 in the FANTASIIA cohort. Events rates were higher in the high-risk category (score ≥3) (1.94%/year vs 0.81%/year in the Murcia cohort; 6.01%/year vs 1.71%/year, in FANTASIIA, both p

This paper considers naive and wild bootstrap procedures to construct pointwise confidence intervals for a nonparametric regression function when the predictor is of functional nature and when the data are dependent. Assuming $\alpha$-mixing conditions on the sample, the asymptotic validity of both procedures is obtained. A simulation study shows promising results when finite sample sizes are used, while an application to electricity demand data illustrates its usefulness in practice.