Test of Randomness of Residuals for the Four-parameter Logistic model used in Obtaining the IC50 Value for Allivum sativum Methanolic Extract Against Aeromonas hydrophila

Abstract

Numerous publications ignore statistical diagnosing of the nonlinear model utilized, and the data might be nonrandom- an essential necessity for all of the parametric statistical evaluation approaches. In cases where the diagnostic tests demonstrate that the residuals reveal a pattern, then a variety of remedies for example nonparametric analysis or shifting to another model should cure the problem. The subject of this study is test for the randomness of the residual for the Four-parameter Logistic model used in obtaining the IC50 Value for Allivum sativum methanolic extract against Aeromonas hydrophila using the Wald–Wolfowitz runs test. The result shows that the number of runs was 10, the expected number of runs under the assumption of randomness was 5.8, indicating the series of residuals had adequate runs. As the p-value was greater than 0.05, the null hypothesis is not rejected demonstrating no substantial evidence that the residuals were nonrandom, and the residuals represent noise.