Constructing Matched Samples with Evidence Factors

2021 ◽  
pp. 205-220
Author(s):  
Paul R. Rosenbaum
Keyword(s):  
Matched Samples

scholarly journals Non-random sampling and association tests on realized returns and risk proxies

2021 ◽  
Author(s):  
Frank Ecker ◽  
Jennifer Francis ◽  
Per Olsson ◽  
Katherine Schipper
Keyword(s):  
Random Sampling ◽  
Reference Sample ◽  
Positive Association ◽  
Cost Of Equity ◽  
Association Tests ◽  
Random Samples ◽  
Distribution Matching ◽  
Matched Samples ◽  
Data Requirements ◽  
Selection Of

AbstractThis paper investigates how data requirements often encountered in archival accounting research can produce a data-restricted sample that is a non-random selection of observations from the reference sample to which the researcher wishes to generalize results. We illustrate the effects of non-random sampling on results of association tests in a setting with data on one variable of interest for all observations and frequently-missing data on another variable of interest. We develop and validate a resampling approach that uses only observations from the data-restricted sample to construct distribution-matched samples that approximate randomly-drawn samples from the reference sample. Our simulation tests provide evidence that distribution-matched samples yield generalizable results. We demonstrate the effects of non-random sampling in tests of the association between realized returns and five implied cost of equity metrics. In this setting, the reference sample has full information on realized returns, while on average only 16% of reference sample observations have data on cost of equity metrics. Consistent with prior research (e.g., Easton and Monahan The Accounting Review 80, 501–538, 2005), analysis using the unadjusted (non-random) cost of equity sample reveals weak or negative associations between realized returns and cost of equity metrics. In contrast, using distribution-matched samples, we find reliable evidence of the theoretically-predicted positive association. We also conceptually and empirically compare distribution-matching with multiple imputation and selection models, two other approaches to dealing with non-random samples.

scholarly journals X-ray properties of He ii λ 1640 emitting galaxies in VANDELS

2020 ◽  
Vol 496 (3) ◽  
pp. 3796-3807
Author(s):  
A Saxena ◽  
L Pentericci ◽  
D Schaerer ◽  
R Schneider ◽  
R Amorin ◽  
...  
Keyword(s):  
Formation Rate ◽  
Galactic Nuclei ◽  
X Ray ◽  
Star Forming ◽  
Set Constraints ◽  
Significant Difference ◽  
Low Metallicity ◽  
Matched Samples ◽  
X Ray Binaries

ABSTRACT We explore X-ray emission from a sample of 18 He ii λ1640 emitting star-forming galaxies at z ∼ 2.3–3.6 from the VANDELS survey in the Chandra Deep Field South, to set constraints on the role of X-ray sources in powering the He ii emission. We find that 4 He ii emitters have tentative detections with S/N ∼ 2 and have X-ray luminosities, LX = 1.5−4.9 × 1041 erg s−1. The stacked luminosity of all 18 He ii emitters is 2.6 × 1041 erg s−1, and that of a subset of 13 narrow He ii emitters (FHWM(He ii) < 1000 km s−1) is 3.1 × 1041 erg s−1. We also measure stacked LX for non-He ii emitters through bootstrapping of matched samples, and find LX = 2.5 × 1041 erg s−1, which is not significantly different from LX measured for He ii emitters. The LX per star formation rate for He ii emitters (log (LX/SFR) ∼ 40.0) and non-emitters (log (LX/SFR) ∼ 39.9) are also comparable and in line with the redshift evolution and metallicity dependence predicted by models. Due to the non-significant difference between the X-ray emission from galaxies with and without He ii, we conclude that X-ray binaries or weak or obscured active galactic nuclei are unlikely to be the dominant producers of He ii ionizing photons in VANDELS star-forming galaxies at z ∼ 3. Given the comparable physical properties of both He ii emitters and non-emitters reported previously, alternative He ii ionizing mechanisms such as localized low-metallicity stellar populations, Pop-III stars, etc. may need to be explored.

Investigating oral fluid and exhaled breath as alternative matrices for anti-doping testing: Analysis of 521 matched samples

2019 ◽  
Vol 176 ◽  
pp. 112810
Author(s):  
Geoffrey D. Miller ◽  
Ryan M. Van Wagoner ◽  
Benjamin J. Bruno ◽  
Jacob D. Husk ◽  
Matthew N. Fedoruk ◽  
...  
Keyword(s):  
Oral Fluid ◽  
Exhaled Breath ◽  
Alternative Matrices ◽  
Matched Samples

scholarly journals Clinical Validation of a Volumetric Absorptive Micro-Sampling Device for Pharmacokinetic Studies With Tranexamic Acid

2021 ◽  
Vol 12 ◽  
Author(s):  
Stanislas Grassin-Delyle ◽  
Elodie Lamy ◽  
Michaela Semeraro ◽  
Iléana Runge ◽  
Jean-Marc Treluyer ◽  
...  
Keyword(s):  
Tranexamic Acid ◽  
Whole Blood ◽  
Plasma Concentrations ◽  
High Sensitivity ◽  
Pharmacokinetic Studies ◽  
Limits Of Agreement ◽  
Sampling Device ◽  
Average Bias ◽  
Alternative Routes ◽  
Matched Samples

We assessed the accuracy of tranexamic acid (TXA) concentrations measured in capillary whole blood using volumetric absorptive micro-sampling (VAMS) devices. Paired venous and VAMS capillary blood samples were collected from 15 healthy volunteers participating in a pharmacokinetic study of alternative routes (oral, IM and IV) of administering TXA. To assess accuracy across a range of concentrations, blood was drawn at different times after TXA administration. We measured TXA concentrations in plasma, whole blood from samples collected by venepuncture and whole blood from venous and capillary samples collected using VAMS devices. TXA was measured using a validated high sensitivity liquid chromatography - mass spectrometry method. We used Bland-Altman plots to describe the agreement between the TXA concentrations obtained with the different methods. In the 42 matched samples, the mean plasma TXA concentration was 14.0 mg/L (range 2.6–36.5 mg/L) whereas the corresponding whole blood TXA concentration was 7.7 mg/L (range 1.6–17.5 mg/L). When comparing TXA concentrations in VAMS samples of venous and capillary whole blood, the average bias was 0.07 mg/L (lower and upper 95% limits of agreement: −2.1 and 2.2 mg/L respectively). When comparing TXA concentrations in venous whole blood and VAMS capillary whole blood, the average bias was 0.7 mg/L (limits of agreement: −2.7 and 4.0 mg/L). Volumetric absorptive micro-sampling devices are sufficiently accurate for use in pharmacokinetic studies of tranexamic acid treatment in the range of plasma concentrations relevant for the assessment of fibrinolysis inhibition.

scholarly journals Repeating the same mistakes: A century of research interested in the effects of bilingualism on non-linguistic cognitive functions

2020 ◽  
Author(s):  
Jennifer Mattschey
Keyword(s):  
Cognitive Development ◽  
Adverse Effects ◽  
Executive Functioning ◽  
20Th Century ◽  
Cognitive Functions ◽  
Cognitive Abilities ◽  
Psychological Research ◽  
Intelligence Tests ◽  
Background Factors ◽  
Matched Samples

At the beginning of the 20th Century, researchers became increasingly interested in the effects of bilingualism on cognitive development. With the emergence of the first standardised intelligence tests, it quickly became clear that bilinguals of all ages performed worse on them than their monolingual peers. Bilingualism was subsequently considered to be detrimental to non-verbal cognitive development. In these early studies, poorly matched samples of bilinguals and monolinguals repeatedly produced results suggesting adverse effects of bilingualism on non-linguistic cognitive functions due to unrelated background factors and/or unsuitable tests. This interest in how bilingualism affects non-linguistic cognitive abilities has continued through time, reflecting trends in psychological research. Researcher addressed its effect on meta-cognition and meta-linguistics in the 1970s and 1980s, before the focus shifted towards executive functioning in the 1990s. However, recent research suggests we may be repeating the same mistakes: poorly matched groups, no accounting for background factors, and inappropriate tasks.

Abstract WP414: Dabigatran Has the Lowest Risk of Hemorrhage and Death Among Patients Taking SSRI/SNRIs That Require Anticoagulation

Stroke ◽  
2017 ◽  
Vol 48 (suppl_1) ◽  
Author(s):  
Salman Al Jerdi ◽  
Mary Sarrazin ◽  
Enrique Leira
Keyword(s):  
Vital Role ◽  
Major Hemorrhage ◽  
Stroke Incidence ◽  
Risk Of Death ◽  
Post Stroke Depression ◽  
Artery Disease ◽  
New Diagnosis ◽  
Risk Of Hemorrhage ◽  
The Impact ◽  
Matched Samples

Introduction: Antidepressants have a vital role in treating post-stroke depression and improving motor recovery. SSRI/SNRI’s have antiplatelet effects, possibly increasing the risk of hemorrhage when taken with anticoagulants. The impact of concomitant SSRI/SNRI use on patients taking different anticoagulants for atrial fibrillation is not known. Methods: Data included Medicare Part A, B, and D claims for 2010-2013. Patients 66 years or older, with a new diagnosis of AF who initiated warfarin, dabigatran (150 mg bid) or rivaroxaban (20 mg od) within 90 days of diagnosis were included. SSRI/SNRI use was identified using prescription claims within +/- 90 days of anticoagulation initiation. Patients taking warfarin, dabigatran, and rivaroxaban with concurrent SSRI/SNRI were matched using a 3-way propensity algorithm. The relative hazards of death, stroke, GI hemorrhage, and other major hemorrhage were compared between anticoagulants. Results: We identified10,773 patients with SSRI/SNRI use, and 37,278 without SSRI/SNRI use. At baseline, patients who received SSRI/SNRIs were more likely than other patients to have several comorbid conditions, including coronary artery disease, liver disease, kidney disease, previous hemorrhage, and previous stroke. In matched samples of patients taking SSRI/SNRI, dabigatran users had the lowest incidence of any hemorrhage, with Hazard Ratio [HR]=0.78 (p=0.07) relative to warfarin; 0.68 (p=0.01) relative to rivaroxaban. Non-GI-hemorrhage was less likely with both dabigatran and rivaroxaban, compared to warfarin, while GI hemorrhage was most likely with rivaroxaban. The risk of death was significantly lower with dabigatran (HR 0.60; p<0.001) and rivaroxaban (HR 0.67; p=0.007), compared to warfarin. There was no difference in stroke incidence by anticoagulant type. Conclusion: In patients with AF and concomitant SSRI/SNRI use, dabigatran has the lowest risk of hemorrhage and mortality.

Collision Detection on Industrial Robots in Repetitive Tasks Using Modified Dynamic Time Warping

Robotica ◽  
2019 ◽  
Vol 38 (10) ◽  
pp. 1717-1736
Author(s):  
Zaviša Gordić ◽  
Kosta Jovanović
Keyword(s):  
Collision Detection ◽  
Dynamic Time Warping ◽  
Reference Signal ◽  
Detection Algorithm ◽  
Industrial Robots ◽  
Absolute Difference ◽  
Time Warping ◽  
Online Application ◽  
Dynamic Time ◽  
Matched Samples

SUMMARYThis paper presents a non-model-based collision detection algorithm for robots without external sensors and with closed control architecture. A reference signal of repetitive motion is recorded from the robot operation. To detect collisions, the reference is compared with measurements from the robot. One of the key contributions is a novel approach to optimal matching of compared signals, which is ensured by the newly developed modified Dynamic Time Warping (mDTW) method presented in this paper. One of the main improvements of the mDTW is that it enables comparing a signal with the most similar section of the other signal. Partial matching also enables online application of time warping principles and reduces the time and computation resources needed to perform matching. In addition to mDTW, two complementary decision rules are developed to identify collisions. The first rule, based on the absolute difference between compared matched samples, uses statistically determined thresholds to perform rapid detection of unambiguous collisions. The second rule is based on Eigen values of the covariance matrix of matched samples, and it employs its higher sensitivity to detect collisions with lower intensity. Results from experimental validation of the proposed collision algorithm on two industrial robots are shown.

Sign in / Sign up

Export Citation Format

Share Document