Preprocessing Data from Excel

I have Raw Data in a excel Spreadsheet. I have power, voltage and current according to time. How to I standardized the data. Please assist with steps.

Thank you

[Mathieu NOE]

hello

what do you mean by "standardized " ?

Answer 1

[Kunal Kandhari]

Hello

You can easily do that by reading the file and calling internal Matlab functions

_______________________________________

Formula for standardization:

where

x=data,

μ=mean,

σ=standard deviatoion

Code for reading file:

T = readtable('fileName.csv','NumHeaderLines',1);
% It will load the file in variable T, and remove the headings

For reading file you can refer:

https://in.mathworks.com/help/matlab/ref/readtable.html

Assign variables to columns:

power= T.Var1;
voltage=T.Var2;
current=T.Var3;
time=T.Var4;

standardization data:

power=(power-mean(power))/std(power);
voltage=(voltage-mean(voltage))/std(voltage);
current=(current-mean(current))/std(current);
time=(time-mean(time))/std(time);

For mean and standard deviation you can refer:

https://in.mathworks.com/help/matlab/ref/mean.html

https://in.mathworks.com/help/matlab/ref/std.html

It will standardization your dataset

[Bhavick Singh]

Thank you for your Help. I have standardized the data. Now I need to use the standardized data and perform Dimensionality reduction predominantly PCA and CCA. Could you assist with me with that with steps.

[Kunal Kandhari]

Hi You can refer the following links for implementing PCA and CCA:

For PCA:

https://in.mathworks.com/help/stats/pca.html

For CCA:

https://in.mathworks.com/help/stats/canoncorr.html

The above pages are self explanatory with some piece of code written to help you understand them.

[Bhavick Singh]

Hello, thanks for the reply. Is there any code for Curvilinear component analysis?

[Kunal Kandhari]

see if referring this links can help you:

https://stats.stackexchange.com/questions/135280/using-canonical-correlation-analysis-cca-to-find-matches

https://in.mathworks.com/help/stats/canoncorr.html