I want to perform moment-matching, and one key second moment would be the ones related with the nominal interest rate, set by the CB through a Taylor Rule (theoretically), which at the same time is the nominal yielding of bonds, i.e. represents the opportunity cost of money.
Thus, I was wondering which would be the appropriate time series for representing this variable, since I have seen for instance in SGU (2004), that they take the nominal interest from the opportunity cost of money using the time series of 3-month treasury bills. In other hand, I’m not sure if it would be more appropriate to use the CB interest target (i.e. the one that is announced, e.g. made from periodic announcements, in the case of the US the Federal Funds Rate), or the actual interest that they target (e.g. Effective Federal Funds Rate)?
A point maybe worth noting is that I’m not working with the US exactly, but with a country with very similar monetary system.