ANOVAN with possibility of matrix as an input (columns as the variables)

11 ビュー (過去 30 日間)
Nuchto
Nuchto 2017 年 11 月 28 日
コメント済み: Nuchto 2017 年 11 月 29 日
Hi, I have my variables in columns in a long matrix (2000x 25), and need to test which of the variables are statistically different. The problem is that Matlab only accepts input in the form of a long response variable which means I need to concatenate the 2000 observations x 25 times, and then specify a "group" vector. Is there a way to perform anovan with a matrix as an input? Thank you! I think this should be the most intuitive and practical way to do anovan in Matlab.

採用された回答

Kaushik Lakshminarasimhan
Kaushik Lakshminarasimhan 2017 年 11 月 28 日
編集済み: Kaushik Lakshminarasimhan 2017 年 11 月 29 日
The complex syntax of anovan allows you to handle the case of unbalanced data. Since you have the same number of observations (2000) for each of the 25 groups, you can use anova1.
If you want to compare each pair of columns, you can use the STATS output of anova1 and perform multiple comparison on that using multcompare.
  1 件のコメント
Nuchto
Nuchto 2017 年 11 月 29 日
Thank you so much! I had missed that function!

サインインしてコメントする。

その他の回答 (0 件)

カテゴリ

Help Center および File ExchangeAnalysis of Variance and Covariance についてさらに検索

タグ

Community Treasure Hunt

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

Start Hunting!

Translated by