Re: Conditions with Statistical Functions
- To: mathgroup at smc.vnet.net
- Subject: [mg66131] Re: [mg66091] Conditions with Statistical Functions
- From: Bob Hanlon <hanlonr at cox.net>
- Date: Sun, 30 Apr 2006 04:21:54 -0400 (EDT)
- Reply-to: hanlonr at cox.net
- Sender: owner-wri-mathgroup at wolfram.com
- Sensitivity: Normal
conditionalMean[x_List?VectorQ, cond_Function]:=
Mean[Select[x,cond]];
conditionalMean[x_List, cond_Function]:=
conditionalMean[#,cond]&/@Transpose[x];
m=Table[Random[Integer,200],{4},{3}]
{{12, 138, 117}, {92, 13, 178}, {147, 131, 123}, {34, 177, 109}}
Mean/@m
{89, 283/3, 401/3, 320/3}
Mean[m]
{285/4, 459/4, 527/4}
conditionalMean[#,(#>100)&]&/@m
{255/2, 178, 401/3, 143}
conditionalMean[m,(#>100)&]
{147, 446/3, 527/4}
Bob Hanlon
---- Gregory Lypny <gregory.lypny at videotron.ca> wrote:
> Yes, thanks Bob. But what about a matrix, where Mean operates column-
> wise? Is there a compact statement where the mean will be calculated
> over the values that exceed 100 in each column all in one shot? What
> I've been doing is looping Select across the columns using Do and
> applying Mean on each pass. I just thought there might be something
> more efficient, if not more elegant.
>
> Regards,
>
> Greg
>
> On Sat, Apr 29, 2006, at 6:53 AM, Bob Hanlon wrote:
>
> > x={10,20,150,200,250};
> >
> > Mean[Select[x,#>100&]]
> >
> > 200
> >
> >
> > Bob Hanlon
> >
>