The hydroxyl number is an experimentally determined one. When an epoxy resin is made by the reaction of bisphenol A with epichlorohydrin, the chemist can control the average degree of polymerization as well as what sort of end groups will be present in the epoxy. Obviously, the use of a slight excess of bisphenol A will lead to a number of terminal (OH) groups attached to benzene rings "which are acidic" in the final product. This number can vary from one case to another.
When -COOH groups (of a polyester, provided that they are terminal) come closer to terminal -OH groups (of an epoxy), the reaction of them is NOT a facile one since catalysis & heating have to be applied to eliminate water.
In my opinion, a better method is needed to find out the amount of (OH) groups present as the one suggested in a previous answer. In any case, one can expect a small hydroxyl number since the condensation polymerization, for making epoxy resins, usually uses equimolar amounts of the 2 monomers "which indicates the presence of few OH groups at the terminals".
When -COOH reacts with epoxy one ester group and one -OH group is generated. This will be the basis of theoretical calculation. 1 Equivalent epoxy will react with one equivalent of carboxyl.