Abstract
Performance estimation denotes a task of estimating the loss that a predictive model will incur on unseen data. These procedures are part of the pipeline in every machine learning task and are used for assessing the overall generalisation ability of models. In this paper we address the application of these methods to time series forecasting tasks. For independent and identically distributed data the most common approach is cross-validation. However, the dependency among observations in time series raises some caveats about the most appropriate way to estimate performance in these datasets and currently there is no settled way to do so. We compare different variants of cross-validation and different variants of out-of-sample approaches using two case studies: One with 53 real-world time series and another with three synthetic time series. Results show noticeable differences in the performance estimation methods in the two scenarios. In particular, empirical experiments suggest that cross-validation approaches can be applied to stationary synthetic time series. However, in real-world scenarios the most accurate estimates are produced by the out-of-sample methods, which preserve the temporal order of observations.

Abstract

Abstract
In time series analysis, Autoregressive Moving Average (ARMA) models play a central role. Because of the importance of parameter estimation in ARMA modeling and since it is based on aggregate time series so often, we analyze the effect of temporal aggregation on estimation accuracy. We derive the relationships between the aggregate and the basic parameters and compute the actual values of the former from those of the latter in order to measure and compare their estimation accuracy. We run a simulation experiment that shows that aggregation seriously worsens estimation accuracy and that the impact increases with the order of aggregation.

Abstract
Purpose: The aim of this study is to validate the current applicability of arteriovenous access banding in high flow access (HFA) and/or haemodialysis access-induced distal ischaemia (HAIDI). Methods: This retrospective study was conducted at the GEV (Grupo de Estudos Vasculares) vascular access centre. The clinical records of consecutive patients undergoing banding for HAIDI and HFA symptoms, between June 2011 and January 2015, were reviewed until April 2015. All vascular access patients' consultation records and surgical notes were reviewed. We analysed and compared patients' age, gender, comorbidities, symptoms and intraoperative ultrasound control. We defined technical failure as recurrence of symptoms, requiring new banding. Excessive banding, access thrombosis, rupture and false aneurysm development were registered as complications. Primary clinical success was defined as improvement of symptoms or effective flow reduction after banding, with no need for reintervention. If one reintervention was necessary, we have defined it as secondary clinical success. Results: Overall, 119 patients underwent banding: 64 (54%) with HAIDI and 55 (46%) with HFA. The HAIDI group was significantly older (65 +/- 13 years compared with 56 +/- 22 years, p = 0.001) and had significantly greater number of patients with diabetes (56% vs 24%, p = 0.004). Primary success was achieved in 85 patients (71.4%) and the secondary success rate was 84.9%. Older age (p = 0.016) and intraoperative ultrasound control (p = 0.012) were significantly associated with primary success. Conclusions: Our results do not corroborate the high incidence of thrombosis previously reported as associated with AV access banding and suggest that ultrasound control is crucial for preventing technical failure. The procedure was effective on both compared groups.

Abstract
Dysfunction problems with vascular access are a concern to patients and dialysis units. The vascular surgeon should analyse such dysfunction and perform a careful assessment of the vascular network in order to find new fistula layouts. We introduce and discuss the case of creation of a radiocephalic fistula with outflow into the forearm basilic vein through rotation of the forearm basilic vein toward the cephalic vein in the forearm of an 88-year-old hemodialysis male patient. This technique enables extending fistula patency and improves cost efficiency.

Abstract
End stage renal disease (ESRD) patients should be educated to maintain and preserve the arteriovenous fistula (AVF) in the best condition. The purpose of this work was to evaluate self-care frequency and factors that influenced such frequency. A prospective study was performed in 101 hemodialysis patients. Self-care behaviors were measured with the Scale of Assessment of Self-Care Behaviours with Arteriovenous Fistula in Hemodialysis. A regression model was used to determine the relevant predictors of self-care frequency and their influence. The incidence of self-care behaviors was 71.0%. The regression model showed that self-care behaviors were positively influenced by gender (female), ESRD etiology (hypertension, polycystic kidneys and other kidney diseases), duration of AVF and negatively by the existence of previous AVF and health professional (doctor). The frequency of self-care behaviors was lower than expected and below an appropriate standard. Education programs designed to improve self-care behaviors with AVF should be further explored in a prospective randomized trial.