Robust Bootstrap Research Articles

R package RobGARCHBoot

This package provides a robust bootstrap procedure to obtain forecast densities for both return and volatilities in a GARCH context. The forecast densities are useful to obtain forecast intervals as well as to estimate risk measures such as Value-at-Risk (VaR). Additionally, we also provide the robust GARCH estimator of Boudt et al. (2013) with the modification introduced by Trucios et at. (2017). This procedure showed good finite sample properties in both Monte Carlo experiments and empirical data. For a recent implementation of this procedure see Trucios (2019).

Phylogenetic relationships of Coreanomecon (Papaveraceae: Chelidonioideae) inferred from seed morphology and nrITS sequence data

The phylogenetic position of the Coreanomecon, a monotypic genus endemic to Korea, has been controversial for a long time. Coreanomecon has variously been placed in its own genus or combined with Hylomecon or Chelidonium. The main purpose of this study was to examine the phylogenetic position of Coreanomecon in relation to genera of the subfamily Chelidonioideae using seed morphology and molecular data. The seed morphology of 10 genera of Chelidonioideae was examined using microtome sections and scanning electron microscopy. The shape and size of exotestal cells varied between genera. The exotestal cells were tangentially elongated in Chelidonium majus, Stylophorum diphyllum and Hylomecon vernalis, whereas the cells were sub‐orbicular and broad in Coreanomecon hylomeconoides. The endotesta was represented by thick palisade‐like cells that contain small rectangular crystals near their outer walls. The mesotesta was totally collapsed in Co. hylomeconoides but wholly or partly persistent in Ch. majus and H. vernalis. The seed surface of Co. hylomeconoides was well ornamented with a unique echinate seed surface, whereas other genera showed plain and reticulate seed surfaces. A phylogenetic analysis of the genera including Coreanomecon was conducted with nuclear ribosomal internal transcribed spacer (nrITS) sequences using genetic distance, maximum parsimony, maximum likelihood and Bayesian inference methods. The results confirm that Co. hylomeconoides is separated from both Hylomecon species and is a sister group to Chelidonium majus and Stylophorum diphyllum with robust bootstrap support. In addition to differences in the absence of cauline leaves and rhizomes, the presence of a hairy stem and leaves, and 12‐pericolpate pollen, seed characters and molecular data strongly support the recognition of Coreanomecon as an independent genus, distinct from Hylomecon and Chelidonium.

Head-to-head comparison of plasma cTnI concentration values measured with three high-sensitivity methods in a large Italian population of healthy volunteers and patients admitted to emergency department with acute coronary syndrome: A multi-center study

BackgroundThe study aim is to compare cTnI values measured with three high-sensitivity (hs) methods in apparently healthy volunteers and patients admitted to emergency department (ED) with acute coronary syndrome enrolled in a large multicentre study. MethodsHeparinized plasma samples were collected from 1511 apparently healthy subjects from 8 Italian clinical institutions (mean age: 51.5 years, SD: 14.1 years, range: 18–65 years, F/M ratio:0.95). All volunteers denied chronic or acute diseases and had normal values of routine laboratory tests. Moreover, 1322 heparinized plasma sample were also collected by 9 Italian clinical institutions from patients admitted to ED with clinical symptoms typical of acute coronary syndrome. The reference study laboratory assayed all plasma samples with three hs-methods: Architect hs-cTnI, Access hs-cTnI and ADVIA Centaur XPT methods. Principal Component Analysis (PCA) was also used to analyze the between-method differences among hs-cTnI assays. ResultsOn average, a between-method difference of 31.2% CV was found among the results of hs-cTnI immunoassays. ADVIA Centaur XPT method measured higher cTnI values than Architect and Access methods. Moreover, 99th percentile URL values depended not only on age and sex of reference population, but also on the statistical approach used for calculation (robust non-parametric vs bootstrap). ConclusionsDue to differences in concentrations and reference values, clinicians should be advised that plasma samples of the same patient should be measured for cTnI assay in the same laboratory. Specific clinical studies are needed to establish the most appropriate statistical approach to calculate the 99th percentile URL values for hs-cTnI methods.

A robust Pearson correlation test for a general point null using a surrogate bootstrap distribution.

In this note we present a robust bootstrap test with good Type I error control for testing the general hypothesis H0: ρ = ρ0. In order to carry out this test we use what is termed a surrogate bootstrap distribution. The test was inspired by the studentized permutation for testing H0: ρ = 0, which was proven to be exact in certain scenarios and asymptotically correct overall. We show that the bootstrap based test is robust to a variety of distributional scenarios in terms of proper Type I error control.

A New Robust Bootstrap Algorithm for the Assessment of Common Set of Weights in Performance Analysis

The performance of the units is defined as the ratio of the weighted sum of outputs to the weighted sum of inputs. These weights can be determined by data envelopment analysis (DEA) models. The inputs and outputs of the related (Decision Making Unit) DMU are assessed by a set of the weights obtained via DEA for each DMU. In addition, the weights are not generally common, but rather, they are very close to zero or they are even equal to zero. This means that some major criteria will not be considered. Another problem is the similarity of the efficiency scores of efficient DMUs. However, this is not the case in reality, and the performance of the DMUs should be completely ranked. Using common weights can solve these problems completely during measuring the performance of DMUs. There are some articles in the literature to determine common weight sets (CSWs), but none of them takes into account the bootstrap approach. This paper introduces a novel, empirical and robust algorithm based on bootstrapping technique to find CSWs.

Statistical inference of random field auto-correlation structure from multiple sets of incomplete and sparse measurements using Bayesian compressive sampling-based bootstrapping

Abstract Bootstrapping is a simulation-based method to provide statistical inference, or to measure accuracy, on estimators of interest, such as random field parameters, i.e., mean and variance of a physical quantity at each location within a random field, and auto-correlation coefficient of the quantities among any two different locations of the field. Statistical inference may be performed using analytical methods together with multiple sets of complete measurement data. However, measurement data may be incomplete sometimes, due to, for examples, sensor failure, storage capacity in collecting measurement data, or cost/difficulty in increasing measurement density. In such a case, it is challenging to perform statistical inference of auto-correlation coefficients, particularly between locations where no data are measured directly. This study aims to develop an innovative and robust bootstrap method to perform statistical inference on auto-correlation coefficient in one dimensional random fields, e.g., time series or variations of a quantity along a spatial direction such as depth, from multiple sets of incomplete and sparse measurements. The proposed approach is based on a Bayesian compressive sampling (BCS)-Karhunen–Loeve (KL) expansion random field generator, which has been used to generate random field samples from sparse measurement data. The proposed bootstrap method is illustrated and validated through both numerical examples and real measurement data. The results obtained from the proposed bootstrap method using multiples sets of incomplete measurement data are consistent with those obtained from analytical methods using multiple sets of complete measurement data.

Inference When There is a Nuisance Parameter under the Alternative and Some Parameters are Possibly Weakly Identified

We present a new robust bootstrap method for a test when there is a nuisance parameter under the alternative, and some parameters are possibly weakly or non-identified. We focus on a Bierens (1990)-type conditional moment test of omitted nonlinearity for convenience, and because of difficulties that have been ignored to date. Existing methods include the supremum p-value which promotes a conservative test that is generally not consistent, and test statistic transforms like the supremum and average for which bootstrap methods are not valid under weak identification. We propose a new wild bootstrap method for p-value computation by targeting specific identification cases. We then combine bootstrapped p-values across polar identification cases to form an asymptotically valid p-value approximation that is robust to any identification case. The wild bootstrap does not require knowledge of the covariance structure of the bootstrapped processes, whereas Andrews and Cheng's (2012, 2013, 2014) simulation approach generally does. Our method allows for robust bootstrap critical value computation as well. Our bootstrap method (like conventional ones) does not lead to a consistent p-value approximation for test statistic functions like the supremum and average. We therefore smooth over the robust bootstrapped p-value as the basis for several tests which achieve the correct asymptotic level, and are consistent, for any degree of identification. A simulation study reveals possibly large empirical size distortions in non-robust tests when weak or non-identification arises. One of our smoothed p-value tests, however, dominates all other tests by delivering accurate empirical size and comparatively high power.

The Sanguicolous Apostome Metacollinia luciensis Jankowski 1980 (Colliniidae, Apostomatia, Ciliophora) Is Not Closely Related to Other Sanguicolous Apostomes.

The apostome family Colliniidae includes species that are adapted to the hemocoel/blood of various invertebrates, particularly crustaceans. To explore the phylogeny of these sanguicolous apostomes, Metacollinia luciensis was collected in August 2015 at Roscoff from the amphipod host, Orchestia gammarellus. Ciliates were Protargol stained and DNA was extracted. The small subunit rRNA (SSUrRNA) and cytochrome c oxidase subunit I (cox1) genes were amplified. Molecular phylogenetic analyses of the SSUrRNA genes unambiguously grouped M.luciensis with other apostomes with robust bootstrap support, but separated it distinctly from the pseudocolliniid clade. While there are only cox1 sequences for a subset of these apostomes, M.luciensis was also distant from the pseudocolliniids and separated from them by species of the exuviotrophic apostome Hyalophysa. These results confirm the distinctness of the families Colliniidae and Pseudocolliniidae.

Robust bootstrap control charts for percentiles based on model selection approaches

Abstract This paper presents two model selection approaches, namely the random data-driven approach and the weighted modeling approach, to construct robust bootstrap control charts for process monitoring of percentiles of the shape-scale class of distributions under model uncertainty. The generalized exponential, lognormal and Weibull distributions are considered as candidate distributions to establish the proposed process control procedures. Monte Carlo simulations are conducted with various combinations of the percentiles, false-alarm rates and sample sizes to evaluate the performance of the proposed robust bootstrap control charts in terms of the average run lengths. Simulation results exhibit that the two proposed robust model selection approaches perform well when the underlying distribution of the quality characteristic is unknown. Finally, the proposed process monitoring procedures are applied to two data sets for illustration.

Bootstrap LM tests for higher-order spatial effects in spatial linear regression models

This paper first extends the methodology of Yang (J Econom 185:33–59, 2015) to allow for non-normality and/or unknown heteroskedasticity in obtaining asymptotically refined critical values for the LM-type tests through bootstrap. Bootstrap refinements in critical values require the LM test statistics to be asymptotically pivotal under the null hypothesis, and for this we provide a set of general methods for constructing LM and robust LM tests. We then give detailed treatments for two general higher-order spatial linear regression models: namely the $$\mathtt{SARAR}(p,q)$$ model and the $$\mathtt{MESS}(p,q)$$ model, by providing a complete set of non-normality robust LM and bootstrap LM tests for higher-order spatial effects, and a complete set of LM and bootstrap LM tests robust against both unknown heteroskedasticity and non-normality. Monte Carlo experiments are run, and results show an excellent performance of the bootstrap LM-type tests.

Robust inference for seemingly unrelated regression models

Seemingly unrelated regression models generalize linear regression models by considering multiple regression equations that are linked by contemporaneously correlated disturbances. Robust inference for seemingly unrelated regression models is considered. MM-estimators are introduced to obtain estimators that have both a high breakdown point and a high normal efficiency. A fast and robust bootstrap procedure is developed to obtain robust inference for these estimators. Confidence intervals for the model parameters as well as hypothesis tests for linear restrictions of the regression coefficients in seemingly unrelated regression models are constructed. Moreover, in order to evaluate the need for a seemingly unrelated regression model, a robust procedure is proposed to test for the presence of correlation among the disturbances. The performance of the fast and robust bootstrap inference is evaluated empirically in simulation studies and illustrated on real data.

Robust bootstrap densities for dynamic conditional correlations: implications for portfolio selection and Value-at-Risk

ABSTRACTMany financial decisions such as portfolio allocation, risk management, option pricing and hedge strategies are based on the forecast of the conditional variances, covariances and correlations of financial returns. Although the decisions depend on the forecasts covariance matrix little is known about effects of outliers on the uncertainty associated with these forecasts. In this paper we analyse these effects on the context of dynamic conditional correlation models when the uncertainty is measured using bootstrap methods. We also propose a bootstrap procedure to obtain forecast densities for return, volatilities, conditional correlation and Value-at-Risk that is robust to outliers. The results are illustrated with simulated and real data.

Is Stellar Hedge Fund Performance for Real?

We apply a robust bootstrap to evaluate the performance of a large universe of hedge funds. Our bootstrap estimates indicate that the performance of the top hedge funds cannot be attributed to chance alone. This is true even after adjusting for back fill bias, serial correlation, and structural breaks. Also, we find that hedge fund alpha differences persist over three year horizons. However, an investment strategy designed around this will run into difficulties as the persistence is often confined to small funds that are effectively closed to new inflows. Moreover, Bayesian estimates suggest that standard alphas may be overestimated by 41% for the average top fund.

Stratification of populated areas using a method of robust principal components

The Provincial Institute of Statistics and Censuses (IPEC) of Santa Fe, Argentina, intended to develop a sampling design that would allow inferences for different levels of disaggregation inside the province. This paper attempts to solve the stratification problem of the populated areas of the prov ince which are the primary sampling units (PSU) of the sampling design. The first classification is related to size: Cities (more than 10,000 inhabitants), towns (between 2,000 and 10,000 inhabitants), and villages (below 2,000 inhabitants). Within each group, the objective is to create homogeneous groups with respect to social and demographical variables from the 2010 Census. Therefore, 10 indicators are considered in order to identify critical needs of the populated areas and an index of critical needs is obtained using a Robust Principal Component Analysis (RPCA) that considers the Bootstrap inference through the Fast and Robust Bootstrap (FRB). Consequently, this index of critical needs is calculated for each PSU and then the Geometric Stratification Method (GSM) is applied in each size stratum to create homogeneous groups. It is observed that the populated areas in the north of the province have greater critical needs than those areas in the south of the province.

Efficiency-solvency linkage of Indian general insurance companies: a robust non-parametric approach

The present study explored the efficiency-solvency linkage in the context of the Indian general insurance sector through a two stage analysis. In the first stage, a robust bootstrap approach is followed for the estimation of technical efficiency scores of 16 in-sample public and private sector general insurance companies. In the second stage, the impact of solvency on the efficiency of the insurers has been considered via truncated regression. The results point out a statistically significant relationship between input/output oriented efficiency and solvency ratio.

Basic Psychological Needs Satisfaction Mediates the Association between Self-Control Skills and Subjective Well-Being.

Although studies have shown that self-control skills (SCSs) are positively linked to both personal and interpersonal outcomes in adolescent students, studies on the putative mechanisms underlying this relationship are scarce. Drawing on Self-Determination Theory and previous studies, we theorized that the association between students’ SCSs and their subjective well-being (SWB) in school may be mediated by students’ perceived satisfaction of their basic psychological needs for competence, relatedness, and autonomy. The sample consisted of 1576 Israeli adolescent students (54% girls) in grades 10–12 (mean age 16) enrolled in 20 schools. A mediation model was tested with structural equation modeling and a robust bootstrap method for testing indirect effects, controlling for school-level variance. The findings supported the hypothesized model and a post hoc multi-group comparison analysis yielded gender invariance in the model. The findings suggest that both girls and boys with high SCSs may perceive themselves as having greater needs satisfaction in school and consequently higher school-related SWB. Implications for policy and practice are discussed.

Robust Bootstrap Densities for Dynamic Conditional Correlations: Implications for Portfolio Selection and Value-at-Risk

Many financial decisions such as portfolio allocation, risk management, option pricing and hedge strategies are based on the forecast of the conditional variances, covariances and correlations of financial returns. Although the decisions are based on forecasts covariance matrix little is known about effects of outliers on the uncertainty associated with these forecasts. In this paper we analyse these effects on the context of dynamic conditional correlation (DCC) models when the uncertainty is measured using bootstrap methods. We also propose a bootstrap procedure to obtain forecast densities for return, volatilities, conditional correlation and VaR that is robust to outliers. The results are illustrated with simulated and real data.

Enhanced bootstrap method for statistical inference in the ICA model

In this paper, we develop low complexity and stable bootstrap procedures for FastICA estimators. Our bootstrapping techniques allow for performing cost efficient and reliable bootstrap-based statistical inference in the ICA model. Performing statistical inference is needed to quantitatively assess the quality of the estimators and testing hypotheses on mixing coefficients in the ICA model. The developed bootstrap procedures stem from the fast and robust bootstrap (FRB) method [1], which is applicable for estimators that may be found as solutions to fixed-point (FP) equations. We first establish analytical results on the structure of the weighted covariance matrix involved in the FRB formulation. Then, we exploit our analytical results to compute the FRB replicas at drastically reduced cost. The developed enhanced FRB method (EFRB) for FastICA permits using bootstrap-based statistical inference in a variety of applications (e.g., EEG, fMRI) in which ICA is commonly applied. Such an approach has not been possible earlier due to incurred substantial computational efforts of the conventional bootstrap. Our simulation studies compare the complexity and numerical stability of the proposed methods with the conventional bootstrap method. We also provide an example of utilizing the developed bootstrapping techniques in identifying equipotential lines of the brain dipoles from electroencephalogram (EEG) recordings.

Portfolio Management with Drawdown-Based Measures

This article analyzes the portfolio management implications of using drawdown-based measures in allocation decisions. The authors introduce <i>modified conditional expected drawdown</i> (MCED), a new risk measure that is derived from portfolio drawdowns, or peak-to-trough losses, of demeaned constituents. They show that MCED exhibits the attractive properties of coherent risk measures that are present in conditional expected drawdown (CED) but are lacking in the historical maximum drawdown (MDD) commonly used in the industry. This article introduces a robust block bootstrap approach to calculating CED, MCED, and marginal contributions from portfolio constituents. First, the authors show that MCED is less sensitive to sample error than CED and MDD. Second, they evaluate several drawdown-based minimum risk and equal-risk allocation approaches within the large-scale simulation framework of Molyboga and L’Ahelec via a subset of hedge funds in the managed futures space that contains 613 live and 1,384 defunct funds over the 1993–2015 period. The authors find that the MCED-based equal-risk approach dominates the other drawdown-based techniques but does not consistently outperform the simple equal volatility–adjusted approach. This finding highlights the importance of carefully accounting for sample error, as reported by DeMiquel et al., and cautions against overreliance on drawdown-based measures in portfolio management. <b>TOPICS:</b>Real assets/alternative investments/private equity, portfolio construction, performance measurement

Robust bootstrap estimates in heteroscedastic semi-varying coefficient models and applications in analyzing Australia CPI data

ABSTRACTThis article deals with the estimation of the parametric component, which is of primary interest, in the heteroscedastic semi-varying coefficient models. Based on the bootstrap technique, we present a procedure for estimating the parameters, which can provide a reliable approximation to the asymptotic distribution of the profile least-square (PLS) estimator. Furthermore, a bootstrap-type estimator of covariance matrix is developed, which is proved to be a consistent estimator of the covariance matrix. Moreover, some simulation experiments are conducted to evaluate the finite sample performance for the proposed methodology. Finally, the Australia CPI dataset is analyzed to demonstrate the application of the methods.