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: [email protected] [mailto:[email protected]] 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 ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
