Hi all,<br>Although the logic of what Jack says is valid, I think the current implementation is a bit confusing--basically because it&#39;s unexpected.  I can&#39;t think of another program that does this, though I&#39;m sure there must be some.  The others that I use (R, Stata, Gauss) don&#39;t. <br>
<br>Since the coding of dummy variable is arbitrary (e.g., you could use 1 -1) why should zero be treated differently from all other scalars and real numbers?  I think it could lead to some unexpected results (for those who don&#39;t dive into the manual).  For the sake of consistency I would prefer that all numbers be treated alike.  <br>

<br>On the other hand if the zero*NA were treated as the exception and  handled with a special command, that would make more logical sense--at least to me.  I find the current implementation is kind of confusing, especially for gretl&#39;s main audience (non programmers like me) ans students coming from other programs that treat the interaction likewise (e.g., Stata and R).   Just my 2 cents, but I think &#39;m with Berend on this one....<br>

<br>Cheers,<br>Lee<br clear="all"><br>-- <br>Lee Adkins<br>Professor of Economics<br><a href="mailto:lee.adkins@okstate.edu" target="_blank">lee.adkins@okstate.edu</a><br><br><a href="http://learneconometrics.com" target="_blank">learneconometrics.com</a><br>