I'm having a slight rounding issue on a formula which is deriving an industry's percentage of a total security type.

 

My information is grouped by Security Type (equity or debt), Region (north america), and industry. The formula gives the percentage which each industry has in the overall composition of the security type.

 

For example --

 

Equity Investment

North America

Health Care and Social Assistance -- 50.00%

Information -- 50.00%

 

Total Equity Investment -- 100%

 

The above is a simple example, in my data I have hundreds of data points within each security type. I'm running into a really small rounding error when summing the formula at the security type level. The formula works on the industry, but is off by a few decimals within the security type grouping. What I'm trying to do is find the variance and plug it back into one of the industries so it foots. I feel like that would cause some kind of infinite loop issue though. Anyone have any suggestions?

 

Would be greatly appreciated!

 

Thanks,
Brandon