Limits...
ESPRESSO: taking into account assessment errors on outcome and exposures in power analysis for association studies.

Gaye A, Burton TW, Burton PR - Bioinformatics (2015)

Bottom Line: They often result in a potentially substantial overestimation of the actual power.In this article, we describe the Estimating Sample-size and Power in R by Exploring Simulated Study Outcomes tool that allows assessment errors in power calculation under various biomedical scenarios to be incorporated.We also report a real world analysis where we used this tool to answer an important strategic question for an existing cohort.

View Article: PubMed Central - PubMed

Affiliation: School of Social and Community Medicine, University of Bristol, UK and.

No MeSH data available.


Graphical view of the GLM analysis in ESPRESSO. After each simulation run a dataset of observed values is generated analysed and the beta coefficient, standard error and z-statistic stored
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4528636&req=5

btv219-F2: Graphical view of the GLM analysis in ESPRESSO. After each simulation run a dataset of observed values is generated analysed and the beta coefficient, standard error and z-statistic stored

Mentions: Steps 2, 3 and 4 are repeated for a number of times equal to the number of runs specified at step 1. After each run, as shown graphically in Figure 2, a matrix, D, of observed data is generated and analysed by GLM, and the estimates (beta, standard error and z-score), obtained from the GLM fit, are stored in three distinct vectors.Fig. 2.


ESPRESSO: taking into account assessment errors on outcome and exposures in power analysis for association studies.

Gaye A, Burton TW, Burton PR - Bioinformatics (2015)

Graphical view of the GLM analysis in ESPRESSO. After each simulation run a dataset of observed values is generated analysed and the beta coefficient, standard error and z-statistic stored
© Copyright Policy - creative-commons
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4528636&req=5

btv219-F2: Graphical view of the GLM analysis in ESPRESSO. After each simulation run a dataset of observed values is generated analysed and the beta coefficient, standard error and z-statistic stored
Mentions: Steps 2, 3 and 4 are repeated for a number of times equal to the number of runs specified at step 1. After each run, as shown graphically in Figure 2, a matrix, D, of observed data is generated and analysed by GLM, and the estimates (beta, standard error and z-score), obtained from the GLM fit, are stored in three distinct vectors.Fig. 2.

Bottom Line: They often result in a potentially substantial overestimation of the actual power.In this article, we describe the Estimating Sample-size and Power in R by Exploring Simulated Study Outcomes tool that allows assessment errors in power calculation under various biomedical scenarios to be incorporated.We also report a real world analysis where we used this tool to answer an important strategic question for an existing cohort.

View Article: PubMed Central - PubMed

Affiliation: School of Social and Community Medicine, University of Bristol, UK and.

No MeSH data available.