Comments on: Standard Deviation, Root Mean Square and the Central Moments (Statistics in Erlang part 3) http://fullof.bs/standard-deviation-root-mean-square-and-the-central-moments-statistics-in-erlang-part-3/ He just never stops talking Tue, 22 Jun 2010 16:27:39 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: John Haugeland http://fullof.bs/standard-deviation-root-mean-square-and-the-central-moments-statistics-in-erlang-part-3/comment-page-1/#comment-868 John Haugeland Mon, 23 Mar 2009 19:29:16 +0000 http://fullof.bs/?p=314#comment-868 There aren't default names for the fifth and sixth central moments, as far as I know. If you find some, let me know, and I'll wrap them. There aren’t default names for the fifth and sixth central moments, as far as I know. If you find some, let me know, and I’ll wrap them.

]]> By: hir http://fullof.bs/standard-deviation-root-mean-square-and-the-central-moments-statistics-in-erlang-part-3/comment-page-1/#comment-867 hir Mon, 16 Mar 2009 08:45:28 +0000 http://fullof.bs/?p=314#comment-867 hello, Would you please tell me what the standard moments (stage 5 & 6) show about a specific distribution?for example B4 (standard moment stage 4) showes the kurtosis and B3 showes the skewness.Please tell me what B5 & B6 show. thanks hello,
Would you please tell me what the standard moments (stage 5 & 6) show about a specific distribution?for example B4 (standard moment stage 4) showes the kurtosis and B3 showes the skewness.Please tell me what B5 & B6 show.
thanks

]]> By: Anders Nygren http://fullof.bs/standard-deviation-root-mean-square-and-the-central-moments-statistics-in-erlang-part-3/comment-page-1/#comment-866 Anders Nygren Mon, 11 Aug 2008 04:19:58 +0000 http://fullof.bs/?p=314#comment-866 To follow up on my previous comment. Like this <code> stddev(L) -> {SqSum, Sum, Len} = lists:foldl(fun (X, {K,S,N}) -> {K+X*X,S+X,N+1} end, {0,0,0}, L), Mean = Sum/Len, math:sqrt(SqSum/Len-Mean*Mean). </code> To follow up on my previous comment. Like this


stddev(L) ->
{SqSum, Sum, Len} = lists:foldl(fun (X, {K,S,N}) ->
{K+X*X,S+X,N+1}
end, {0,0,0}, L),
Mean = Sum/Len,
math:sqrt(SqSum/Len-Mean*Mean).

]]> By: Anders Nygren http://fullof.bs/standard-deviation-root-mean-square-and-the-central-moments-statistics-in-erlang-part-3/comment-page-1/#comment-865 Anders Nygren Mon, 11 Aug 2008 03:17:02 +0000 http://fullof.bs/?p=314#comment-865 Statistics are an area with little support in erlang at the moment so I think that this is a good initiative. But You need to be more careful with how many times You traverse the lists of values. For instance in std_deviation You traverse the list 5 times. I seem to remember from my statistics course, (>25 years ago), that it is possible to calculate stddev with only one traversal. Statistics are an area with little support in erlang at the moment so I think that this is a good initiative.
But You need to be more careful with how many times You traverse the lists of values.
For instance in std_deviation You traverse the list 5 times. I seem to remember from my statistics course, (>25 years ago), that it is possible to calculate stddev with only one traversal.

]]>