Hello,

I am working with a dataset that contains dollar amounts that range from $0 to several million. When importing the dataset, Stata properly chooses the right variable type to preserve precision, so I have a mix of float, long, and double. My question is whether the
Code:
 collapse
command will know if it is passing into territory where precision may become an issue and update the variable type for the outputted variables. For example, will it know that a large total of smaller floats may add up to a very large number, and switch to double in the output?

I checked the collapse documentation but did not see anything.

Thank you for your help.