achieves_planned_objective logit-log curve fitting A logit-log curve fitting is a curve fitting where first the limits y_0 an y_infty of y when x->0 and x->infinity, respectively, are estimated from the input data points (x_1, y_1), (x_2,y_2), ..., (x_n, y_n). Then a curve with equation log((y-y_0)/(y_infty-y))=a+b log(x) is obtained, where a and b are determined to optimize its fit to the input data points. ARTICLE: Plikaytis B.D. et al. (1991), J. Clin. Microbiol. 29(7): 1439-1448 Elisabetta Manduchi James Malone Melanie Courtot Ryan Brinkman The above definition refers to the 'fully specified' logit-log model. The reduced form of this, when it is assumed that y_0=0, is named 'partially specified' logit-log model. Typically used in an enzyme-linked immunosorbent assay (ELISA) to model the relationship between optical density (OD) and dilution. In this case OD_0 (also referred to OD_min) and OD_infty (also referred to OD_max) correspond to the theoretical OD of the assay at zero and infinite concentrations, respectively. logit-log curve fitting curve fitting data transformation curve fitting objective ready for release