XLatexIt! run report...
*** Found expression $\begin{ matrix } 0 \\ { y }_{ i,1 } \\ . \\ . \\ . \\ { y }_{ i,T-2 } \end{ matrix }$
LaTeX process returned 1
Proceeding anyway...
*** Found expression $ar1, $
Image was already generated
Thank you Allin for ur quick response!
I'll write below.

01.07.2013 04:17, Allin Cottrell:
On Sun, 30 Jun 2013, Pindar wrote:


I think that accessors for the various test-statistics and 
p-values would be useful like $ar1, $ar2.

We're contemplating a more general way of making "special" model 
data available. Watch this space.
Cool! I gonna follow this development :-)

      

        
Have you already thought about incorporating a 'collapse' switch?


      

      
Which would do what, exactly?


    

    It's a means to limit instrument proliferation.

    According to Roodman (2006) "This embodies the same expectation but
    conveys slightly less information, since it generates a single
    moment condition" (p.23)

    

    In footnote 11 on page 24: "After conceiving of such instrument sets
    and adding a collapse option to xtabond2, I discovered precedents.
    Adapting Arellano and Bond's (1998) dynamic panel package, DPD for
    Gauss, and performing System GMM, Calderon, Chong, and Loayza
    (2002) use such instruments, followed by Beck and Levine (2004) and
    Carkovic and Levine (2005). Roodman (2009) demonstrates the
    superiority of collapsed instruments in some common situations with
    simulations."

    

    Here is the new first column after collapsing:

    

    $\begin{ matrix }
      0 \\ { y }_{ i,1 } \\ . \\ . \\ . \\ { y }_{ i,T-2 } \end{ matrix
      }$

    

    

    

    

      

      

        
The GMM() function does not support the use of lists right now. 
Could this be improved?


      

      
Maybe, but it seems this might encourage the over-proliferation of 
instruments that produces singular "A" matrices and hence somewhat 
indeterminate results.

    

    True. I now use the springf with @nlags and three GMM() to test
    down.

    

      

Allin Cottrell


_______________________________________________
Gretl-users mailing list
Gretl-users@lists.wfu.edu
http://lists.wfu.edu/mailman/listinfo/gretl-users