How can I use the PCA function to find the principle components of a variable?

Question

Anna Vaculik 2018-10-17

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/424419-how-can-i-use-the-pca-function-to-find-the-principle-components-of-a-variable

回答： Bernhard Suhm 2018-11-9

Hello, I am working on developing a 4D matrix that will have lat,lon,day,time for a months worth of temperature data. Can I simply say z=pca(goes_lst)? To break down the components within? Or is there a better way? Thank you!

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Bernhard Suhm 2018-11-9

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/424419-how-can-i-use-the-pca-function-to-find-the-principle-components-of-a-variable#answer_346080

You'll get the loadings for your variables. But think about whether applying PCA this way to your data makes sense. We'll expect time of day to explain a fair amount of the variability in temperature, the day will just reflect the weather patterns of that particular month (or the absence thereof). Latitude and longitude would explain climate differences in geographically quite separate areas. Think about what kind of model are you trying to develop, and whether or not you need preprocessing for that.