Error using @regress fuction

Hello, I am trying to create a function as follows with a set of variables that I get from a table named Change1:

X = [Change1.Ones Change1.ReturnsMarket]
func = @regress(Change1.ReturnsCleaned,X);

However, everytime I get the following error message:

Invalid expression. When calling a function or indexing a variable, use parentheses. Otherwise, check for
mismatched delimiters.

Thank you all for your help !

Answer 1

[Star Strider]

Try this:

func = @(Q)regress(Q,X);

then call it as:

[b,bint] = func(Change1.ReturnsCleaned);

(It seems a bit inefficient to me to wrap regress in an anonymous function. Personal opinion only.)

Answer 2

[Walter Roberson]

The @ operator can be used in four related syntaxes:

@NAME : create handle to function indicated by the name
@NAME1.NAME2 : create handle to function whose name inside its Package is as given
@() EXPRESSION : create handle to anonymous function with no parameters that evaluates to the given expression when invoked 
@(NAME LIST) EXPRESSION : create handle to anonymous function with dummy parameters named inside the ()

You however have @EXPRESSION which is not valid syntax.

@ always has to do with creating handles, except possibly for some class definition purposes. If you want to invoke regress function instead of creating a handle then leave out the @

Answer 3

[Nabil Benhadda]

It is because I am trying to use it as part of the splitapply function, can I do so without the @?

[Star Strider]

To use it with splitapply, I would just use:

func = @(Q)regress(Q,X);

[Nabil Benhadda]

Thank you a lot !

However now I have an issue, I use the code as following:

Change1.Groups = findgroups(Change1.Numero);

X = [Change1.Groups Change1.Ones Change1.ReturnsMarket];

Q = Change1.ReturnsCleaned;

func = @(Q)regress(Q,X);

Regress = splitapply(func,X,Q);

I get the following error message:

Error using splitapply (line 61)

Group numbers must be a vector of positive integers, and cannot be a sparse vector.

Error in Untitled (line 8)

Regress = splitapply(func,X,Q);

[Star Strider]

I cannot help you with that. I only provided the correct function syntax in my Answer.

[Nabil Benhadda]

Thank you a lot !

[Star Strider]

As always, my pleasure!

[Walter Roberson]

Q = Change1.ReturnsCleaned;

You are using that as your grouping variable, but it is not obvious to us that it contains only positive integers.

Change1.Groups = findgroups(Change1.Numero);
X = [Change1.Groups Change1.Ones Change1.ReturnsMarket];

findgroups does return the kind of positive integers needed for a grouping variable, but you are folding that information in as part of X and using X as the variable to be split rather than as the grouping variable. That seems odd.