Description Usage Arguments Value Author(s) References
Grid representation by this method is an alternative to representing distribution of the market, such as importance sampling or stratified sampling.However, these techniques focus on sub-domains of the distribution, and thus, in order to apply such methods, we must forego the full flexibility on the specification of the views. A different approach, which preserves the generality of the views specification, consists in selecting the scenarios x_j deterministically as a grid and then associate with each of them the suitable probability p_j ( integrated over I_j ). The the generic interval _j contains the j-th point of the grid. Once the grid is defined, the entropy optimization can be applied to replace p_j with the new posterior probabilities to reflect the views. We generate the grid here using the Gauss-Hermite quadrature method.
1 |
J |
a numeric containing the number of points on the grid |
x a matrix containing the zeroes of Hermite polynomials as a function of polynomial degree
Ram Ahluwalia ram@wingedfootcapital.com
A. Meucci - "Fully Flexible Extreme Views". http://ssrn.com/abstract=1542083
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.