I suppose you are referring to array functions from the Analysis library: unfrtunately those functions are limited to treat doubles only. I don't know of any alternative for floats or doubles: for the simpler of them you may want to create your own equivalent in a for loop; for more complex functions you can create a temporary array of doubles and use ConvertArrayType function to exchange data to and from it before and after calling the array function.
Depending on other functions you are using into your application, you can consider promotinig all of your arrays to double and use them: PC are now plenty of memory to use and very fast CPU frequency so it could be a feasible solution.
