It looks like John is pointing in the right direction. Using his hint, I'm
able
to replicate 2 out of 3 of Coyle's other examples:
]pod=. >(1,(%9),%5);9 1 2;5 0.5 1 NB. "Prediction of Dynamics", p. 7
1 0.11111111 0.2
9 1 2
5 0.5 1
(%+/){."1 >2{dgeev_jlapack_ pod
0.066010328 0.61532831 0.31866136
]und=. >1 1 5;1 1 3;(%5),(%3),1 NB. "Understandability", p. 7
1 1 5
1 1 3
0.2 0.33333333 1
(%+/){."1 >2{dgeev_jlapack_ und
0.48064009 0.40538755 0.11397237
]rod=. >(1,(%3),%9);(3,1,%3);9 3 1 NB. "Replication of detail", p. 7
1 0.33333333 0.11111111
3 1 0.33333333
9 3 1
(%+/){."1 >2{dgeev_jlapack_ rod
_0.2 0.3 0.9
NB. This differs... (vs. 0.077 0.231 0.692)
On 3/1/06, John Randall <[EMAIL PROTECTED]> wrote:
>
> John Randall wrote:
> >
> > (b) Coyle's data is weirdly normalized.
> >
> w=:0.232 0.402 0.061 0.305
> +/w
> 1
>
> Best wishes,
>
> John
>
...
--
Devon McCormick
^me^ at acm.
org is my
preferred e-mail
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
