Why would you want to do this though? A variable name is just a handle used in code. Suppose you create these dynamic variable names. Now everywhere afterwards in the code you have to keep working with these and using ugly code constructs to do so because you can't just refer to a variable properly by its name in the code.
Just use a struct if you want to do something like this e.g.
cat.mean = ...
dog.mean = ...
or if you only want one variable rather than separate ones for cat, dog, etc
animal.type = cat;
animal.mean = ...
animal(2).type = 'dog'
animal(2).mean = ...
