Record Display for the EPA National Library CatalogRECORD NUMBER: 3 OF 4
|Fitting Statistical Distributions to Air Quality Data by the Maximum Likelihood Method.
|Holland, D. M. ; Fitz-Simons, T. ;
|Environmental Monitoring Systems Lab., Research Triangle Park, NC.
|Air pollution ; Statistical distributions ; Concentration(Composition) ; Normal density functions ; Weibull density functions ; Probability theory ; Reprints ; Air quality data ; Maximum likelihood estimation ; Lognormal density functions ; Gamma density functions ; Johnson density functions ; Beta density functions ; Goodness of fit ; MAXFIT computer program
A computer program has been developed for fitting statistical distributions to air pollution data using maximum likelihood estimation. Appropriate uses of this software are discussed and a grouped data example is presented. The program fits the following continuous distributions: normal, three-parameter lognormal, three-parameter gamma, three-parameter Weibull, Johnson (S sub B), and four-parameter beta. The parameters of each distribution are estimated by closed solutions and the Nelder-Mead Simplex iterative search. Graphical output contains a plot of the fitted distribution superimposed upon the histogram of the data for each model. Six goodness-of-fit criteria are supplied and ranked by the program to aid in the selection of the most appropriate choice among the six models.