[Jprogramming] Re : Sigma notation

[Linda Alvord]

J has a funny way of taking over how you think about problems. Here's a
triple sigma sort of problem which seems to be an accounting problem.  At
each stage you must have a formula and a domain. The domain is lurking in
the original matrix and the formulas must be applied before you apply each
sigma or  +/ 

   +/10*+/3++/*:i.2 3 5
86000
  
   i.2 3 5
 0  1  2  3  4
 5  6  7  8  9
10 11 12 13 14

15 16 17 18 19
20 21 22 23 24
25 26 27 28 29
   
   +/*:i.2 3 5
225 257 293 333  377
425 477 533 593  657
725 797 873 953 1037
   +/3++/*:i.2 3 5
1384 1540 1708 1888 2080
   +/10*+/3++/*:i.2 3 5
86000

   +/10*+/3++/*:i.5 3 2
85730

Have fun!

Linda


-----Original Message-----
From: programming-bounces at forums.jsoftware.com
[mailto:programming-bounces at forums.jsoftware.com] On Behalf Of Alex
Giannakopoulos
Sent: Saturday, December 22, 2012 12:53 AM
To: J Programming forum
Subject: [Jprogramming] Re : Sigma notation

Brian,

Well, yes, that is basically what my adverb is trying to generalize.
However, I stress that I am not looking at this so much for my own use, but
more in terms of how to present it to a student that has learnt or is
learning sigma notation.

+/ fn (start irange end)

is OK, I suppose, but like I said, I am a little concerned about the
notation.
On the one hand I can see that it simplifies some things, on the other I
wonder if it will generalise well.  How about, for instance, a double or
triple sigma summation?
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm