Define a set of independent variables for fitlm

Suppose I have a table T with labels

visitors price proxy_January proxy_February proxy_March

I want to run a regression

visitors = a + b*price + c_1*proxy_January + ...

A brute force way is to write down

X=[T.price T.proxy_January (then the rest of proxy variables)]

and run

mdl = fitlm(X,T.visitors)

But I would like to have Matlab to write down all the proxy variables instaad of me listing all by hand.

I understand that I can get the list of labels with T.Properties.VariableNames but I need to get the list of labels and data. Please advise.

Answer 1

[Star Strider]

Possibly:

X = T{:,2:end};

then:

mdl = fitlm(X,T.visitors)

If ‘T’ has only numeric variables, that should work. .

[alpedhuez]

Just to know, what is the difference between () and {}?

[Star Strider]

The indexing of table objects derives from cell array indexing.

The () (parentheses) extracts a table variable, and {} (curly braces)extracts the contents of a table variable. This is covered in detail in Access Data in Tables.

[alpedhuez]

Thank you.

[Star Strider]

As always, my pleasure!

(My apologies for the delay in responding.)