st: re: estimate precision in MATA: correction

[Christopher Baum <kit.baum@bc.edu>]

<>
My statement "You can't affect the precision with which computations are performed. " is technically incorrect (thanks, Nick!) because there are various Mata functions which enable you to increase the accuracy with which a computation is performed (e.g., quadcross()). However, those functions take arguments which are doubles and return objects which are doubles. Thus, you do not have a choice about the accuracy of Mata numeric objects, although you do have alternatives in computing them---and quadcross() will be much better than an explicit matrix multiplication for ill-conditioned matrices.

Kit

Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                              An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
   An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/