MathGroup Archive 2006

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Matrix multiplication problem fixed via dot operation.

  • To: mathgroup at smc.vnet.net
  • Subject: [mg64611] Re: [mg64552] Matrix multiplication problem fixed via dot operation.
  • From: Andrzej Kozlowski <akoz at mimuw.edu.pl>
  • Date: Fri, 24 Feb 2006 00:18:07 -0500 (EST)
  • References: <200602221058.FAA23462@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

Does that mean you have been under the impression that multiplication  
of square matrices works without the dot?
To disabuse yourself of this notion try multiplying in this way the  
identity matrix by a non-diagonal matrix, e.g.

IdentityMatrix[2] {{1, 1}, {1, 1}}

and see what you get.

Andrzej Kozlowski





On 22 Feb 2006, at 11:58, Chris Young wrote:

> To answer my own posting, it looks like the "dot" operator is needed
> when the matrices aren't both square.
>
> E.g.: a row vector with components a and b times a column vector with
> components c and d will yield the correct result, the dot-product.
>
> I'm still puzzled why Mathematica doesn't do the same thing without  
> the dot.
>
> Chris Young
>
>> In[3]:=
>> \!\(\*
>>   RowBox[{
>>     RowBox[{"(", GridBox[{
>>           {"a", "b"}
>>           }], ")"}], ".",
>>     RowBox[{"(", GridBox[{
>>           {"c"},
>>           {"d"}
>>           }], ")"}]}]\)
>>
>> Out[3]=
>> {{a c+b d}}
>


  • Prev by Date: Re: finding the position of a pattern in list
  • Next by Date: Re: finding the position of a pattern in list
  • Previous by thread: Re: Matrix multiplication problem fixed via dot operation.
  • Next by thread: Re: Matrix multiplication problem fixed via dot operation.