A Comparison Of Equivalence Testing In Combination With Hypothesis Testing And Effect Sizes

Mecklin, Christopher J.

doi:10.22237/jmasm/1067645160

Cited by 8 publications

(6 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is also confirmed by the metrics in Table I. For each metric, we compared the performance of both the embedded and MATLAB controller and tested statistical equivalence in terms of an equivalence test; see, for example, [45]. Unlike the more common statistical difference tests (like the T-test or the ranksum test), the equivalence test checks the hypothesis that means of the two samples are different; rejection of this hypothesis at some significance level (we consider 0.05) implies equivalence of the two samples with high confidence.…”

Section: Hardware-in-the-loop Experimental Evaluationsupporting

confidence: 55%

Embedded Model Predictive Control for a Wearable Artificial Pancreas

Chakrabarty

Healey

Shi

et al. 2020

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

While artificial pancreas (AP) systems are expected to improve the quality of life among people with type 1 diabetes mellitus (T1DM), the design of convenient systems that optimize the user experience, especially for those with active lifestyles, such as children and adolescents, still remains an open research question. In this work, we introduce an embeddable design and implementation of model predictive control (MPC) of AP systems for people with T1DM that significantly reduces the weight and on-body footprint of the AP system. The embeddable controller is based on a zone MPC that has been evaluated in multiple clinical studies. The proposed embedded zone MPC features a simpler design of the periodic safe zone in the cost function and the utilization of state-of-the-art alternating minimization algorithms for solving the convex programming problems inherent to MPC with linear models subject to convex constraints. Off-line closed-loop data generated by the FDA-accepted UVA/Padova simulator is used to select Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/ publications/rights/index.html for more information.

show abstract

Section: Hardware-in-the-loop Experimental Evaluationsupporting

confidence: 55%

Embedded Model Predictive Control for a Wearable Artificial Pancreas

Chakrabarty

Healey

Shi

et al. 2020

IEEE Trans. Contr. Syst. Technol.

View full text Add to dashboard Cite

show abstract

“…Moreover, Seaman and Serlin (1998) and Tryon (2001) discussed equivalence confidence intervals for two-group comparisons in psychological methods. Mecklin (2003) used equivalence testing in conjunction with standard hypothesis testing and effect sizes, and Cribbie, Gruman, and Arpin-Cribbie (2004) compared the test of equivalence with the traditional Student's t test. The objective of this type of hypothesis testing is to ask whether the test programme can reach the same effect as that of a standard one by setting an a priori critical margin.…”

Section: Introductionmentioning

confidence: 99%

Sample size planning with the cost constraint for testing superiority and equivalence of two independent groups

Guo

Chen

Luh

2011

Brit J Math & Statis

View full text Add to dashboard Cite

The allocation of sufficient participants into different experimental groups for various research purposes under given constraints is an important practical problem faced by researchers. We address the problem of sample size determination between two independent groups for unequal and/or unknown variances when both the power and the differential cost are taken into consideration. We apply the well-known Welch approximate test to derive various sample size allocation ratios by minimizing the total cost or, equivalently, maximizing the statistical power. Two types of hypotheses including superiority/non-inferiority and equivalence of two means are each considered in the process of sample size planning. A simulation study is carried out and the proposed method is validated in terms of Type I error rate and statistical power. As a result, the simulation study reveals that the proposed sample size formulas are very satisfactory under various variances and sample size allocation ratios. Finally, a flowchart, tables, and figures of several sample size allocations are presented for practical reference.

show abstract

“…Equivalence testing is useful when the researcher wants to show that two means are not statistically different and requires the identification of a clinically significant difference (d) between outcomes of competing treatments. In practice, d is often chosen to be a percentage (usually 15%-25%) of the mean of the comparison group (Mecklin, 2002). This number is then used to create a tolerance interval around zero beyond which equivalence would be rejected.…”

Section: Discussionmentioning

confidence: 99%

Early Intervention for Autism With a Parent-Delivered Qigong Massage Program: A Randomized Controlled Trial

Silva

Schalock

Gabrielsen

2011

The American Journal of Occupational Therapy

View full text Add to dashboard Cite

A recent randomized controlled trial (RCT) of a dual parent and trainer-delivered qigong massage intervention for young children with autism resulted in improvement of measures of autism as well as improvement of abnormal sensory responses and self-regulation. The RCT evaluated the effects of the parent-delivered component of the intervention. Forty-seven children were randomly assigned to treatment and wait-list control groups. Treatment group children received the parent-delivered program for 4 mo. Trained therapists provided parent training and support. Improvement was evaluated in two settings-preschool and home-by teachers (blind to group) and parents. Results showed that the parent-delivered program was effective in improving measures of autism (medium effect size) and sensory and self-regulatory responses (large effect size). Teacher data on measures of autism were confirmed by parent data. Results indicate that the parentdelivered component of the program provided effective early intervention for autism that was suitable for delivery at home.

show abstract

A Comparison Of Equivalence Testing In Combination With Hypothesis Testing And Effect Sizes

Cited by 8 publications

References 33 publications

Embedded Model Predictive Control for a Wearable Artificial Pancreas

Embedded Model Predictive Control for a Wearable Artificial Pancreas

Sample size planning with the cost constraint for testing superiority and equivalence of two independent groups

Early Intervention for Autism With a Parent-Delivered Qigong Massage Program: A Randomized Controlled Trial

Contact Info

Product

Resources

About