scholarly journals Integrating informative hypotheses into the EffectLiteR framework

Methodology ◽  
2021 ◽  
Vol 17 (4) ◽  
pp. 307-325
Author(s):  
Caroline Keck ◽  
Axel Mayer ◽  
Yves Rosseel
Keyword(s):  
Null Hypothesis ◽  
Simulated Data ◽  
Wald Test ◽  
Wald Statistic ◽  
Average Effect ◽  
Treatment Groups ◽  
Stochastic Nature ◽  
Classical Test ◽  
Non Linear ◽  
New Treatment

Using the EffectLiteR framework, researchers can test classical null hypotheses about effects of interest via Wald and F-tests, while taking into account the stochastic nature of group sizes. This paper aims at extending EffectLiteR to test informative hypotheses, assuming for example that the average effect of a new treatment is greater than the average effect of an old treatment, which in turn is greater than zero. We present a simulated data example to show two methodological novelties. First, we illustrate how to use the Fbar- and generalized linear Wald test to assess informative hypotheses. While the classical test did not reach significance, the informative test correctly rejected the null hypothesis, indicating the need to take into account the order of the treatment groups. Second, we demonstrate how to account for stochastic group sizes in informative hypotheses using the generalized non-linear Wald statistic. The paper concludes with a short data example.

scholarly journals Independence estimators for re-randomisation trials in multi-episode settings: a simulation study

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Brennan C. Kahan ◽  
Ian R. White ◽  
Sandra Eldridge ◽  
Richard Hooper
Keyword(s):  
Simulation Study ◽  
Treatment Effect ◽  
Standard Errors ◽  
Average Effect ◽  
Treatment Allocation ◽  
Treatment Groups ◽  
Different Types ◽  
Added Benefit ◽  
Nominal Coverage ◽  
New Treatment

Abstract Background Re-randomisation trials involve re-enrolling and re-randomising patients for each new treatment episode they experience. They are often used when interest lies in the average effect of an intervention across all the episodes for which it would be used in practice. Re-randomisation trials are often analysed using independence estimators, where a working independence correlation structure is used. However, research into independence estimators in the context of re-randomisation has been limited. Methods We performed a simulation study to evaluate the use of independence estimators in re-randomisation trials. We focussed on a continuous outcome, and the setting where treatment allocation does not affect occurrence of subsequent episodes. We evaluated different treatment effect mechanisms (e.g. by allowing the treatment effect to vary across episodes, or to become less effective on re-use, etc), and different non-enrolment mechanisms (e.g. where patients who experience a poor outcome are less likely to re-enrol for their second episode). We evaluated four different independence estimators, each corresponding to a different estimand (per-episode and per-patient approaches, and added-benefit and policy-benefit approaches). Results We found that independence estimators were unbiased for the per-episode added-benefit estimand in all scenarios we considered. We found independence estimators targeting other estimands (per-patient or policy-benefit) were unbiased, except when there was differential non-enrolment between treatment groups (i.e. when different types of patients from each treatment group decide to re-enrol for subsequent episodes). We found the use of robust standard errors provided close to nominal coverage in all settings where the estimator was unbiased. Conclusions Careful choice of estimand can ensure re-randomisation trials are addressing clinically relevant questions. Independence estimators are a useful approach, and should be considered as the default estimator until the statistical properties of alternative estimators are thoroughly evaluated.

Test procedure and sample size determination for a proportion study using a double-sampling scheme with two fallible classifiers

2017 ◽  
Vol 28 (4) ◽  
pp. 1019-1043 ◽  
Author(s):  
Shi-Fang Qiu ◽  
Xiao-Song Zeng ◽  
Man-Lai Tang ◽  
Wai-Yin Poon
Keyword(s):  
Sample Size ◽  
Null Hypothesis ◽  
Test Procedure ◽  
Sample Size Determination ◽  
Double Sampling ◽  
Score Statistic ◽  
Wald Statistic ◽  
Test Procedures ◽  
Practical Applications ◽  
Logit Transformation

Double sampling is usually applied to collect necessary information for situations in which an infallible classifier is available for validating a subset of the sample that has already been classified by a fallible classifier. Inference procedures have previously been developed based on the partially validated data obtained by the double-sampling process. However, it could happen in practice that such infallible classifier or gold standard does not exist. In this article, we consider the case in which both classifiers are fallible and propose asymptotic and approximate unconditional test procedures based on six test statistics for a population proportion and five approximate sample size formulas based on the recommended test procedures under two models. Our results suggest that both asymptotic and approximate unconditional procedures based on the score statistic perform satisfactorily for small to large sample sizes and are highly recommended. When sample size is moderate or large, asymptotic procedures based on the Wald statistic with the variance being estimated under the null hypothesis, likelihood rate statistic, log- and logit-transformation statistics based on both models generally perform well and are hence recommended. The approximate unconditional procedures based on the log-transformation statistic under Model I, Wald statistic with the variance being estimated under the null hypothesis, log- and logit-transformation statistics under Model II are recommended when sample size is small. In general, sample size formulae based on the Wald statistic with the variance being estimated under the null hypothesis, likelihood rate statistic and score statistic are recommended in practical applications. The applicability of the proposed methods is illustrated by a real-data example.

scholarly journals Deep learning dark matter map reconstructions from DES SV weak lensing data

2020 ◽  
Vol 492 (4) ◽  
pp. 5023-5029 ◽  
Author(s):  
Niall Jeffrey ◽  
François Lanusse ◽  
Ofer Lahav ◽  
Jean-Luc Starck
Keyword(s):  
Dark Matter ◽  
Deep Learning ◽  
Power Spectrum ◽  
Cosmological Parameters ◽  
Simulated Data ◽  
Wiener Filtering ◽  
Weak Lensing ◽  
Gaussian Shape ◽  
Non Linear ◽  
Mass Mapping

ABSTRACT We present the first reconstruction of dark matter maps from weak lensing observational data using deep learning. We train a convolution neural network with a U-Net-based architecture on over 3.6 × 105 simulated data realizations with non-Gaussian shape noise and with cosmological parameters varying over a broad prior distribution. We interpret our newly created dark energy survey science verification (DES SV) map as an approximation of the posterior mean P(κ|γ) of the convergence given observed shear. Our DeepMass1 method is substantially more accurate than existing mass-mapping methods. With a validation set of 8000 simulated DES SV data realizations, compared to Wiener filtering with a fixed power spectrum, the DeepMass method improved the mean square error (MSE) by 11 per cent. With N-body simulated MICE mock data, we show that Wiener filtering, with the optimal known power spectrum, still gives a worse MSE than our generalized method with no input cosmological parameters; we show that the improvement is driven by the non-linear structures in the convergence. With higher galaxy density in future weak lensing data unveiling more non-linear scales, it is likely that deep learning will be a leading approach for mass mapping with Euclid and LSST.

An Extended Dynamic Model for Kinematic Positioning

2003 ◽  
Vol 56 (1) ◽  
pp. 79-88 ◽  
Author(s):  
Michael Moore ◽  
Jinling Wang
Keyword(s):  
Kalman Filter ◽  
Data Acquisition ◽  
Dynamic Model ◽  
Dynamic Modelling ◽  
Simulated Data ◽  
Acquisition Rate ◽  
Dynamic Conditions ◽  
Kinematic Positioning ◽  
Non Linear ◽  
Extended Dynamic

The main problems faced by a dynamic model within a Kalman filter occur when the system experiences unexpected dynamic conditions, a change in data acquisition rate, or when the dynamics of the system are non-linear. To minimize the errors produced from dynamic modelling in unusual conditions, an extended dynamic model is developed in this paper, and its usefulness demonstrated through comparison of the performance of a Kalman filter's response to simulated data with a standard dynamic model and the extended dynamic model. The results show that, in use, the proposed extended dynamic model is superior to a standard dynamic model, due mainly to its ability to adapt to a wider range of dynamic conditions, which in turn ensures the optimization of the Kalman filter and the consequent generation of reliable positioning results.

High Skp2 Expression Is an Independent Predictor of Unfavorable Outcome in 333 Patients with Diffuse Large B Cell Lymphoma (DLBCL).

Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 1560-1560
Author(s):  
Ritsuko Seki ◽  
Koichi Ohshima ◽  
Fumio Kawano ◽  
Toshihiko Murayama ◽  
Yukiyoshi Moriuchi ◽  
...  
Keyword(s):  
Clinical Stage ◽  
Combined Treatment ◽  
B Cell Lymphoma ◽  
Similar Distribution ◽  
Prognostic Impact ◽  
Advanced Clinical Stage ◽  
Treatment Groups ◽  
Skp2 Expression ◽  
New Treatment

Abstract The addition of rituximab to CHOP (CHOP-R) chemotherapy has resulted in an improved outcome for patients with DLBCL and has recently been shown to diminish the prognostic impact of several recognized biomarkers. S-phase kinase-associated protein 2 (Skp2) is a proto-oncogene that has been shown to be expressed in a number of tumors. We have reported that Skp2 expression in tumor cells is an unfavorable prognostic factor in DLBCL. In the present study, we investigated the significance of Skp2 expression in the patients with DLBCL treated with CHOP or CHOP-R. DLBCL patients (333 cases) were entered into this study, based on the availability of paraffin blocks for interpretable immunohistochemistry for all antigens (CD10, Bcl-6, MUM1, Bcl-2, Skp2). All patients were treated with either CHOP (201) or CHOP-R (132) from 1996 to 2005, and were diagnosed as having DLBCL at the twenty different hospitals. All specimens were histopathologically reviewed before entering into this study. Their clinical characteristics, including either the IPI or R-IPI factors, were evenly matched. The median follow-up of living patients was 3.7 and 2.1 y for CHOP vs CHOP-R, respectively. DLBCL were assigned to GCB subtype (40.8%: 136/333) or non-GCB subtype (59.2%: 197/333) based on the method of Hans et al., Blood 103: 275–82 (2004), with similar distribution in both treatment groups. Expression of bcl-6 (p<0.05) or GCB subtypes (p<0.05) was associated with better overall survival (OS), whereas expression of bcl-2 (p<0.05) was associated with worse OS in CHOP treatment group. The addition of R was associated with an improved survival in the non-GCB subtype and resulted in same as that of GCB subtype. The survival benefit of both low Bcl-2 and high Bcl-6 expressions diminished in combined treatment with R to CHOP. There were 97 patients with high Skp2 expression (>60% positive cells) (97/333: 29.1%). High Skp2 expression was found in both GCB subtype (28.6%) and non-GCB subtype (30.3%). In advanced clinical stage or extranodal involvement (>2), the patients with high Skp2 expression had worse survival than those with low Skp2 expression (p<0.05). Interestingly, in CHOP-R group, high Skp2 expression was the strong biomarker of worse prognosis (p<0.05). DLBCL patients with high Skp2 expression did not benefit from the addition of R to CHOP. Therefore, Skp2 may be a useful prognostic marker in recent rituximab era. The new treatment strategy is necessary for the DLBCL patients with high Skp2 expression. Figure Figure

Is the classical Wald test always suitable under response-adaptive randomization?

2016 ◽  
Vol 27 (8) ◽  
pp. 2294-2311 ◽  
Author(s):  
Alessandro Baldi Antognini ◽  
Alessandro Vagheggini ◽  
Maroussa Zagoraiou
Keyword(s):  
Sample Size ◽  
Power Function ◽  
Wald Test ◽  
Correct Choice ◽  
Consistent Estimator ◽  
Initial Sample ◽  
Normal Response ◽  
Adaptive Randomization ◽  
Classical Test ◽  
The Impact

The aim of this paper is to analyze the impact of response-adaptive randomization rules for normal response trials intended to test the superiority of one of two available treatments. Taking into account the classical Wald test, we show how response-adaptive methodology could induce a consistent loss of inferential precision. Then, we suggest a modified version of the Wald test which, by using the current allocation proportion to the treatments as a consistent estimator of the target, avoids some degenerate scenarios and so it should be preferable to the classical test. Furthermore, we show both analytically and via simulations how some target allocations may induce a locally decreasing power function. Thus, we derive the conditions on the target guaranteeing its monotonicity and we show how a correct choice of the initial sample size allows one to overcome this drawback regardless of the adopted target.

scholarly journals Antidepressant effects of augmentative transcranial magnetic stimulation

2007 ◽  
Vol 191 (5) ◽  
pp. 441-448 ◽  
Author(s):  
Uwe Herwig ◽  
Andreas J. Fallgatter ◽  
Jacqueline Höppner ◽  
Gerhard W. Eschweiler ◽  
Martina Kron ◽  
...  
Keyword(s):  
Transcranial Magnetic Stimulation ◽  
Magnetic Stimulation ◽  
Rating Scales ◽  
Repetitive Transcranial Magnetic Stimulation ◽  
Antidepressant Medication ◽  
Antidepressant Effect ◽  
Double Blind ◽  
Treatment Groups ◽  
New Treatment ◽  
Depressive Episodes

BackgroundRepetitive transcranial magnetic stimulation (rTMS) has been proposed as a new treatment option for depression. Previous studies were performed with low sample sizes in single centres and reported heterogeneous results.AimsTo investigate the efficacy of rTMS as augmentative treatment in depression.MethodIn a randomised, double-blind, sham-controlled multicentre trial 127 patients with moderate to severe depressive episodes were randomly assigned to real or sham stimulation for 3 weeks in addition to simultaneously initiated antidepressant medication.ResultsWe found no difference in the responder rates of the real and the sham treatment groups (31% in each) or in the decrease of the scores on the depression rating scales.ConclusionsThe data do not support previous reports from smaller samples indicating an augmenting or accelerating antidepressant effect of rTMS. Further exploration of the possible efficacy of other stimulation protocols or within selected sub-populations of patients is necessary.

The analysis of Westphalian German Spirantization

Diachronica ◽  
2014 ◽  
Vol 31 (2) ◽  
pp. 223-266 ◽  
Author(s):  
T. Alan Hall
Keyword(s):  
Present Article ◽  
Short Vowel ◽  
Linear Representations ◽  
Crucial Component ◽  
Non Linear ◽  
New Treatment

Westphalian German Spirantization refers to the change from an original prevocalic long vowel to the corresponding short vowel plus fricative (i.e. [ɣ]). For example, the [ɪɣ] sequence in the Westphalian word [klɪɣə] “bran” derived historically from [iː]. The present article offers a new treatment for the historical shift from [iː] to [ɪɣ] — as well as similar ones involving other vowels — which breaks the process down into five separate changes. It is argued that each of these changes modified non-linear representations involving syllables, moras and segmental features. A crucial component of the proposed analysis is that each of the five changes is seen as a repair to a constraint.

scholarly journals A power approximation for the Kenward and Roger Wald test in the linear mixed model

PLoS ONE ◽  
2021 ◽  
Vol 16 (7) ◽  
pp. e0254811
Author(s):  
Sarah M. Kreidler ◽  
Brandy M. Ringham ◽  
Keith E. Muller ◽  
Deborah H. Glueck
Keyword(s):  
Mixed Model ◽  
Linear Mixed Model ◽  
Wald Test ◽  
Small Sample ◽  
Power Calculation ◽  
Wald Statistic ◽  
Power Approximation ◽  
Group Randomized Trial ◽  
Approximate Moments ◽  
Image Position

We derive a noncentral F power approximation for the Kenward and Roger test. We use a method of moments approach to form an approximate distribution for the Kenward and Roger scaled Wald statistic, under the alternative. The result depends on the approximate moments of the unscaled Wald statistic. Via Monte Carlo simulation, we demonstrate that the new power approximation is accurate for cluster randomized trials and longitudinal study designs. The method retains accuracy for small sample sizes, even in the presence of missing data. We illustrate the method with a power calculation for an unbalanced group-randomized trial in oral cancer prevention.

scholarly journals An Asymmetric Investigation of Remittance and Trade Openness Impact on Inequality: Evidence From Selected South Asian Countries

2021 ◽  
Vol 12 ◽  
Author(s):  
Liu Fang ◽  
Md. Qamruzzaman
Keyword(s):  
South Asian ◽  
Unit Root ◽  
Trade Openness ◽  
Wald Test ◽  
Ordinary Least Squares ◽  
Linear Process ◽  
Unit Root Tests ◽  
Asian Countries ◽  
Non Linear ◽  
Bds Test

This study’s motivation is to explore the relationship pattern between remittance, trade openness, and inequality of selected south Asian countries for the 1976–2018 period. The study performed non-linear tests, including unit root tests, non-linearity applying ordinary least squares (OLS) and BDS tests, non-linear autoregressive distributed lagged (NARDL) tests, and asymmetry causality tests to assess their association. Study findings with non-linear unit root tests suggest that the research variables follow the non-linear process of becoming stationary from non-stationary. The non-linear OLS and BDS test results confirm the existence of non-linearity among research variables, implying rejection of the null hypothesis of “no non-linearity.” Furthermore, the results of the Wald test in NARDL confirm the availability of asymmetric links among variables. Besides this, the results of NARDL confirm the long-run asymmetric relationship between remittances, trade openness, and inequality in all sample nations. Findings suggest that both positive and negative shocks in remittances and trade openness is critical to either instituting or vexing the present state of inequality in the economy in the long term. In the directional relationship with asymmetry causality, the study shows that the feedback hypothesis holds to explain the asymmetric causal effects that are positive shocks in remittances and trade openness toward inequality.

Sign in / Sign up

Export Citation Format

Share Document