On Sensitivity of Metalearning: An Illustrative Study for Robust Regression

2020 ◽  
pp. 261-270
Author(s):  
Jan Kalina
Keyword(s):  
Robust Regression

Common Genetic Variants in ABO and CLEC4M Modulate the Pharmacokinetics of Recombinant FVIII in Severe Hemophilia A Patients

2020 ◽  
Vol 120 (10) ◽  
pp. 1395-1406 ◽  
Author(s):  
Iris Garcia-Martínez ◽  
Nina Borràs ◽  
Marta Martorell ◽  
Rafael Parra ◽  
Carme Altisent ◽  
...  
Keyword(s):  
Blood Group ◽  
Half Life ◽  
Robust Regression ◽  
Hemophilia A ◽  
Illumina Miseq ◽  
Activity Levels ◽  
Severe Hemophilia ◽  
Single Nucleotide Variants ◽  
Common Genetic Variants ◽  
Recombinant Fviii

AbstractThe pharmacokinetic (PK) response of severe hemophilia A (HA) patients to infused factor VIII (FVIII) shows substantial variability. Several environmental and genetic factors are associated with changes in FVIII plasma levels and infused FVIII PK. Based on the hypothesis that factors influencing endogenous FVIII can affect FVIII PK, the contribution of single-nucleotide variants (SNVs) in candidate genes was investigated in 51 severe HA patients. The effects of blood group, F8 variant type, von Willebrand factor antigen and activity levels, age, and weight were also explored. The myPKFiT device was used to estimate individual PK parameters, and SNVs and clinically reportable F8 variants were simultaneously analyzed in an Illumina MiSeq instrument, using the microfluidics-based Fluidigm Access Array system. The contribution of SNVs to FVIII half-life and clearance was addressed by robust regression modeling, taking into account other modulators. In line with previous studies, we provide robust evidence that age, body weight, and blood group, as well as SNVs in ABO and CLEC4M, participate in the variability of FVIII PK in HA patients. Main results: each copy of the rs7853989 (ABO) allele increases FVIII half-life by 1.4 hours (p = 0.0131) and decreases clearance by 0.5 mL/h/kg (p = 5.57E-03), whereas each additional rs868875 (CLEC4M) allele reduces FVIII half-life by 1.1 hours (p = 2.90E-05) and increases clearance by 0.3 mL/h/kg (p = 1.01E-03). These results contribute to advancing efforts to improve FVIII replacement therapies by adjusting to each patient's PK profile based on pharmacogenomic data. This personalized medicine will decrease the burden of treatment and maximize the benefits obtained.

Increased sensitivity in neuroimaging analyses using robust regression

NeuroImage ◽  
2005 ◽  
Vol 26 (1) ◽  
pp. 99-113 ◽  
Author(s):  
Tor D. Wager ◽  
Matthew C. Keller ◽  
Steven C. Lacey ◽  
John Jonides
Keyword(s):  
Robust Regression ◽  
Increased Sensitivity

scholarly journals Novel robust time series analysis for long-term and short-term prediction

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Hiroshi Okamura ◽  
Yutaka Osada ◽  
Shota Nishijima ◽  
Shinto Eguchi
Keyword(s):  
Robust Regression ◽  
Simulated Data ◽  
New Method ◽  
Short Term ◽  
Least Absolute Deviations ◽  
Term Prediction ◽  
Regression Approach ◽  
Short Term Prediction ◽  
Long Term Prediction

AbstractNonlinear phenomena are universal in ecology. However, their inference and prediction are generally difficult because of autocorrelation and outliers. A traditional least squares method for parameter estimation is capable of improving short-term prediction by estimating autocorrelation, whereas it has weakness to outliers and consequently worse long-term prediction. In contrast, a traditional robust regression approach, such as the least absolute deviations method, alleviates the influence of outliers and has potentially better long-term prediction, whereas it makes accurately estimating autocorrelation difficult and possibly leads to worse short-term prediction. We propose a new robust regression approach that estimates autocorrelation accurately and reduces the influence of outliers. We then compare the new method with the conventional least squares and least absolute deviations methods by using simulated data and real ecological data. Simulations and analysis of real data demonstrate that the new method generally has better long-term and short-term prediction ability for nonlinear estimation problems using spawner–recruitment data. The new method provides nearly unbiased autocorrelation even for highly contaminated simulated data with extreme outliers, whereas other methods fail to estimate autocorrelation accurately.

scholarly journals Prediction of seasonal patterns of porcine reproductive and respiratory syndrome virus RNA detection in the U.S. swine industry

2020 ◽  
Vol 32 (3) ◽  
pp. 394-400
Author(s):  
Giovani Trevisan ◽  
Leticia C. M. Linhares ◽  
Bret Crim ◽  
Poonam Dubey ◽  
Kent J. Schwartz ◽  
...  
Keyword(s):  
Robust Regression ◽  
Historical Data ◽  
The United States ◽  
Respiratory Syndrome Virus ◽  
Seasonal Patterns ◽  
Test Results ◽  
Swine Industry ◽  
Cyclic Pattern ◽  
Rna Detection ◽  
Virus Rna

We developed a model to predict the cyclic pattern of porcine reproductive and respiratory syndrome virus (PRRSV) RNA detection by reverse-transcription real-time PCR (RT-rtPCR) from 4 major swine-centric veterinary diagnostic laboratories (VDLs) in the United States and to use historical data to forecast the upcoming year’s weekly percentage of positive submissions and issue outbreak signals when the pattern of detection was not as expected. Standardized submission data and test results were used. Historical data (2015–2017) composed of the weekly percentage of PCR-positive submissions were used to fit a cyclic robust regression model. The findings were used to forecast the expected weekly percentage of PCR-positive submissions, with a 95% confidence interval (CI), for 2018. During 2018, the proportion of PRRSV-positive submissions crossed 95% CI boundaries at week 2, 14–25, and 48. The relatively higher detection on week 2 and 48 were mostly from submissions containing samples from wean-to-market pigs, and for week 14–25 originated mostly from samples from adult/sow farms. There was a recurring yearly pattern of detection, wherein an increased proportion of PRRSV RNA detection in submissions originating from wean-to-finish farms was followed by increased detection in samples from adult/sow farms. Results from the model described herein confirm the seasonal cyclic pattern of PRRSV detection using test results consolidated from 4 VDLs. Wave crests occurred consistently during winter, and wave troughs occurred consistently during the summer months. Our model was able to correctly identify statistically significant outbreak signals in PRRSV RNA detection at 3 instances during 2018.

scholarly journals Role of International Trade Competitive Advantage and Corporate Governance Quality in Predicting Equity Returns: Static and Conditional Model Proposals for an Emerging Market

2021 ◽  
Vol 14 (3) ◽  
pp. 125
Author(s):  
Erol Muzir ◽  
Cevdet Kizil ◽  
Burak Ceylan
Keyword(s):  
Corporate Governance ◽  
International Trade ◽  
Dynamic Models ◽  
Robust Regression ◽  
Emerging Market ◽  
Static Model ◽  
Conditional Model ◽  
Portfolio Returns ◽  
Governance Quality ◽  
Corporate Governance Quality

This paper aims to develop some static and conditional (dynamic) models to predict portfolio returns in the Borsa Istanbul (BIST) that are calibrated to combine the capital asset-pricing model (CAPM) and corporate governance quality. In our conditional model proposals, both the traditional CAPM (beta) coefficient and model constant are allowed to vary on a binary basis with any degradation or improvement in the country’s international trade competitiveness, and meanwhile a new variable is added to the models to represent the portfolio’s sensitivity to excess returns on the governance portfolio (BIST Governance) over the market. Some robust and Bayesian linear models have been derived using the monthly capital gains between December 2009 and December 2019 of four leading index portfolios. A crude measure is then introduced that we think can be used in assessing governance quality of portfolios. This is called governance quality score (GQS). Our robust regression findings suggest both superiority of conditional models assuming varying beta coefficients over static model proposals and significant impact of corporate governance quality on portfolio returns. The Bayesian model proposals, however, exhibited robust findings that favor the static model with fixed beta estimates and were lacking in supporting significance of corporate governance quality.

scholarly journals Robust regression with compositional covariates

2021 ◽  
pp. 107315
Author(s):  
Aditya Mishra ◽  
Christian L. Müller
Keyword(s):  
Robust Regression

Percentile transformation and recalibration functions allow harmonization of thyroid-stimulating hormone (TSH) immunoassay results

2020 ◽  
Vol 58 (10) ◽  
pp. 1663-1672 ◽  
Author(s):  
Andrea Padoan ◽  
Aldo Clerico ◽  
Martina Zaninotto ◽  
Tommaso Trenti ◽  
Renato Tozzoli ◽  
...  
Keyword(s):  
Robust Regression ◽  
Reference Intervals ◽  
Thyroid Stimulating Hormone ◽  
Validation Dataset ◽  
New Approach ◽  
Coefficients Of Variation ◽  
Significant Difference ◽  
Before And After ◽  
Roche Diagnostics ◽  
Stimulating Hormone

AbstractBackgroundThe comparability of thyroid-stimulating hormone (TSH) results cannot be easily obtained using SI-traceable reference measurement procedures (RPMs) or reference materials, whilst harmonization is more feasible. The aim of this study was to identify and validate a new approach for the harmonization of TSH results.MethodsPercentile normalization was applied to 125,419 TSH results, obtained from seven laboratories using three immunoassays (Access 3rd IS Thyrotropin, Beckman Coulter Diagnostics; Architect System, Abbott Diagnostics and Elecsys, Roche Diagnostics). Recalibration equations (RCAL) were derived by robust regressions using bootstrapped distribution. Two datasets, the first of 119 EQAs, the second of 610, 638 and 639 results from Access, Architect and Elecsys TSH results, respectively, were used to validate RCAL. A dataset of 142,821 TSH values was used to derive reference intervals (RIs) after applying RCAL.ResultsAccess, Abbott and Elecsys TSH distributions were significantly different (p < 0.001). RCAL intercepts and slopes were −0.003 and 0.984 for Access, 0.032 and 1.041 for Architect, −0.031 and 1.003 for Elecsys, respectively. Validation using EQAs showed that before and after RCAL, the coefficients of variation (CVs) or among-assay results decreased from 10.72% to 8.16%. The second validation dataset was used to test RCALs. The median of between-assay differences ranged from −0.0053 to 0.1955 mIU/L of TSH. Elecsys recalibrated to Access (and vice-versa) showed non-significant difference. TSH RI after RCAL resulted in 0.37–5.11 mIU/L overall, 0.49–4.96 mIU/L for females and 0.40–4.92 mIU/L for males. A significant difference across age classes was identified.ConclusionsPercentile normalization and robust regression are valuable tools for deriving RCALs and harmonizing TSH values.

Robust SURE estimates of profitability in the Egyptian insurance market

2021 ◽  
pp. 1-13
Author(s):  
Ahmed H. Youssef ◽  
Amr R. Kamel ◽  
Mohamed R. Abonazel
Keyword(s):  
Robust Regression ◽  
Ordinary Least Squares ◽  
Seemingly Unrelated Regression ◽  
Insurance Market ◽  
Insurance Companies ◽  
Regression Equations ◽  
Robust Estimators ◽  
Exogenous Variables ◽  
Net Profit ◽  
Mm Estimation

This paper proposed three robust estimators (M-estimation, S-estimation, and MM-estimation) for handling the problem of outlier values in seemingly unrelated regression equations (SURE) models. The SURE model is one of regression multivariate cases, which have especially assumption, i.e., correlation between errors on the multivariate linear models; by considering multiple regression equations that are linked by contemporaneously correlated disturbances. Moreover, the effects of outliers may permeate through the system of equations; the primary aim of SURE which is to achieve efficiency in estimation, but this is questionable. The goal of robust regression is to develop methods that are resistant to the possibility that one or several unknown outliers may occur anywhere in the data. In this paper, we study and compare the performance of robust estimations with the traditional non-robust (ordinary least squares and Zellner) estimations based on a real dataset of the Egyptian insurance market during the financial year from 1999 to 2018. In our study, we selected the three most important insurance companies in Egypt operating in the same field of insurance activity (personal and property insurance). The effect of some important indicators (exogenous variables) issued by insurance corporations on the net profit has been studied. The results showed that robust estimators greatly improved the efficiency of the SURE estimation, and the best robust estimation is MM-estimation. Moreover, the selected exogenous variables in our study have a significant effect on the net profit in the Egyptian insurance market.

Sign in / Sign up

Export Citation Format

Share Document