scientific article; zbMATH DE number 3986496
In psychology one sometimes studies two-parameter families of distribution functions V(z/x,y) where x and y are the parameters. These distribution functions often depend on x and y only through a real-valued function F such that \(V(z/x,y)=V_ 1(z/F(x,y))\). A very important class of models related to Rasch's theory of specific objectivity [see e.g., \textit{G. Rasch}, Proc. 4th Berkeley Sympos. math. Statist. Probab. 4, 321- 333 (1961; Zbl 0107.368)] are models where F is of the form \(F(x,y)=f(g(x)+h(y))\). Functions of this form will be called latent additive. It is shown that the differential equation \[ F_{xy}(F_{xx}F^ 2_ y-F_{yy}F^ 2_ x)=F_ xF_ y(F_{xxy}F_ y-F_{xyy}F_ x) \] nearly characterizes the class of latent additive functions. (The full result is given in corollary 1 at the end of the paper.)
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 Q3750006)