I am using Excel 2010 and getting different RMSE (root mean square error) values when I calculate then myself (in Excel) vs relying on the Data Analysis Regression tool (where it is called the "standard error" in the output table). The differences are only noticeable when I have a relatively low sample size, but even at high sample sizes, the number is still different at the decimal level.
I have read that Excel calculates some statistical values incorrectly (especially in graphs), but I cannot find a source stating that RMSE values are calculated incorrectly.
Can someone confirm that this is also a problem for them using Excel 2010? Is there a solution for correcting Excel's handling of the RMSE calculation via the Data Analysis tool? Do newer versions of Excel still have this problem?
I have read that Excel calculates some statistical values incorrectly (especially in graphs), but I cannot find a source stating that RMSE values are calculated incorrectly.
Can someone confirm that this is also a problem for them using Excel 2010? Is there a solution for correcting Excel's handling of the RMSE calculation via the Data Analysis tool? Do newer versions of Excel still have this problem?
