Comparative Analysis of Astrophysical Data by Different Nonlinear Regression Strategies

In this paper we study opportunities provided by some of the well-known nonlinear regression strategies in order to analyze available astrophysical data. Astrophysical data are currently of special interest due to many factors complicating the application of modern statistical methods of analysis. On the other hand, it is very challenging from perspectives of predicting possible cosmological events that will happen in the future, as well as those that have happen in the past, i.e. before we started to observe the universe. For the analysis of available data set we test four methods: weighted nonlinear regression; robust nonlinear regression; least-square nonlinear regression. Results of numerical analysis are discussed.