I'm trying to do a t-test in Excel but I'm not sure how to use it properly. I got seven different methods for calculating a company's value. For each method, I do an estimation of the value and therefore get an (absolute) deviation from the actual value in percent (so zero is the best estimation and the higher the deviation gets, the worse the result). The dataset comprises 300 companies. My objective is to compare each of the seven methods pairwise against each other to find out which is the best one.
Which is the right t-test in the data analysis add-in in excel? There are three different t-tests (Paired Two Sample for Means, Two-Sample Assuming Equal Variances, Two-Sample assuming unequal Variances). What are the differences? What assumption regarding the variances can I make?
Thank you!
