Illustration of comparison between the actual bivariate joint PDF surface with that of a conventional classical marginal distribution approach approximation showing the discrete set of contour curves visualized by three dimensional lines that are used to construct joint PDF surface, and that of a scattered data approximation of the corresponding normalized bivariate surface. In (a) the actual continuous probability density function distribution is shown which must be modelled and approximated through a visualization in a three dimensional space where the x coordinate corresponds to the random variable , the y coordinate corresponds to the random variable ξ_λ and the z coordinate corresponds to the joint PDF . In (b) the modelling and approximation is shown where the three dimensional surface is approximated by a series of three dimensional lines within the three dimensional space. In (c) the modelling and approximation is visualized for the equivalent surface that is built up from discrete three dimensional scattered points.