scientific article; zbMATH DE number 3920199

From MaRDI portal
Publication:3694993





[For the entire collection see Zbl 0571.00016.] For nonlinear parametric programming problems \(P(\lambda)=\min \{f(x,\lambda):\) \(x\in M(\lambda)\}\), \(\lambda\in \Lambda\), the structures in the parameter space \(\Lambda\) have been proved to be very complicated and inconvenient for usual applications. A method is presented in the paper for the illustrative case of quadratic programming which allows to find structures in \(\Lambda\) being easy to survey also for some classes of nonlinear problems (in particular for problems with an objective function which is a linear function of their coefficients). To simplify the discussion, problems with fixed feasible convex solution sets M in Euclidean spaces are considered. The domains of local stationarity are described and a classification of stationary points of local and global optima is given. The procedure is illustrated by an example.











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 Q3694993)