Abstract
Introduction: The presence of outliers appears as an unavoidable main problem in experimental studies. Outliers can greatly distort parameter estimates and subsequent standard errors. Consequently, inferences about the parameters are misleading. In this case, applying outlier robust statistical procedures should be considered. Robust estimation often relies on a dispersion function that is more slowly varying at large values than the square function. However, the choice of tuning constant in dispersion functions may impact the estimation efficiency to a great extent. Methods: The data used in this paper is part of a concentration–response study that shows the contraction of corpus cavernosum induced by phenylephrine in the organ bath. Because of the existence of an outlier to achieve robust estimations, M-estimation method and Huber function as a dispersion function are used. Three different measures for tuning constant were considered. (0.1, 1.5, 4) Results: Based on the negative log likelihood, robust model had the best fit when . In addition, because of the presence outlier, the Population Average [PA] in common model considerably underestimates the mean response in the upper asymptote. As result, using Huber function when in the robust model to apply the data was led to these results, cumulative administration of phenylephrine (0.1µM - 300µM) caused concentration-dependent contractions in strips of rat corpus cavernosum (-Log EC 50 was 5 ± 0.31, 95% CI= 5.92 to 4.21). Conclusions: To estimate parameters of the model because of existence of an outlier in dataset, M-estimation method and Huber function as a dispersion function has been applied. The appropriate choice of tuning constant can be led to accurate results.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call