Determining Reference Definitions

So, defining what a reference is is, for me, the current hurdle I'm attempting to overcome to create a functioning economic system (details found here and  here).  To recap, references are the relative importance of the various trade goods - goods with more references are more valuable than those with fewer.  Essentially, each reference is equivalent in value to each other reference, so having 9 references of bees and only 2 of gold means that bees are 4.5 times more important than gold.  Each reference is associated with a quantity of the good it represents, which allows us to see the relative production values of that good.

Now, since Alexis' world is Earth c. 1650, he is able to approximate the global production values of each of his goods from the global production values of goods today, then divide those totals by the total number of references for each good to get an exact (ish) quantity of goods for each reference.  It is an excellent system, but one that absolutely does not work for the world of Prodigy, which has but 800,000 square miles of land mass currently inhabited by humans (slightly less than the size of the Louisiana Purchase, or about the size of Mozambique).  Due to the region's small size, compared to the planet as a whole, and the fact that the region does not correspond to any geographic area on Earth, I have to go a different route.

In Alexis' post linked above, he suggests setting the reference value of gold to 2,000 oz per reference.  So, since I am unwilling to arbitrarily define these references, let's use that value to define our whole system.  Each reference is equal in value to each other reference, so by using the current economy, we can get at least some prices on the table.  Yes, the availability of resources in today's society is drastically different from medieval availability.  Yes, many of these goods are commodities which fluctuate widely in price over time (gold, especially).  So, there are several sizable problems with this methodology, but I've spent a while trying to come up with a better one and have yet to find it.

So, let's find the cash value of 2,000oz of gold, and then the price per large unit (mostly tons and bushels, since I'm ignoring the livestock for now) for each of my raw goods.  By dividing the cash value of a reference of gold  by the price per large unit, I'll get the number of large units equal in value to 1 reference of gold.  Aha!  We have reference values!

Now for the next problem: price correction.  I have approximate values of raw goods, and I also have a price list of various goods from the medieval era (here), so by using the latter to get a correction to the former, I ought to be able to come up with a rebalanced set of prices.  We'll see how it goes.