How correlated list of stats

3 次查看（过去 30 天）
Chris
Chris 2011-8-29
I have the stats for about 100 players. Each stat sheet is for a given player for games played. I am trying to find a function that will compare the stats for the 100 players and find out which ones are the closest in terms of correlation. Preferably percentage correlated. What function would be best for this.
  1 个评论
the cyclist
the cyclist 2011-8-29
There is not really enough detail here to do anything but guess at what stats you have, or what you mean by "closest in terms of correlation". I suggest you do two things to help us answer your question. First, post a sample of the data. Second, spend a little more time thinking, conceptually, about what you really mean by "closest".

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

pangyuteng
pangyuteng 2011-8-29
Cyclist is right, but I'd still like to give it a shot base on your question. Assuming all parameters are numbers, say you have 5 parameters for each player, with a total of 10 players.
statsheet = [magic(5),magic(5)];
You can simply obtian the correlation coefficient between each player with the following.
[r,p]=corrcoef(statsheet);
preferablyPercentageCorrelated = r.^2;
'preferablyPercentageCorrelated' will be the "preferably percentage correlated" values between each player.

更多回答（0 个）

类别

在 Help Center 和 File Exchange 中查找有关 Just for fun 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by