scientific article; zbMATH DE number 4050810

From MaRDI portal
Publication:3787326





[For the entire collection see Zbl 0634.00012.] Some aspects of the mathematical description of special growth processes are presented. Empirical models for the quantitative description of the processes were developed and an adaption to the measured time series has to be found. The so-called hyperlogistic model is used as a class of admissible model functions. In many cases, the measured values consist of one time series only. Then it is not possible to realize exactly the weighted least squares parameter estimation. Therefore, the time series is interpreted as realization of a continuous stochastic process. It is demonstrated that one can construct Markov diffusion processes the expectation functions of which are hyperlogistic growth functions. Furthermore, the Markov property offers a chance to realize the maximum likelihood estimation of expectation and covariance function simultaneously out of only one time series.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3787326)