Applying a Factor Times Face Value to Determine Market Value?

ChuckC

I'm curious to know if anyone has come up with some sort of factor that when multiplied by the face value of one's coin collection equals the approximate market value/price that the set would sell for in today. For example, I calculated the face value of my coin collection to be $264.68 and perhaps it's worth 4 times that amount or maybe it's worth 8 times that amount. Maybe it's worth only twice that. Some of the variables that one would have to take into consideration are:

-Coin type
-Coin condition
-Coin scarcity
-Demand
-Precious metal content & current spot price
-Frequency of coins shown to sell for hundreds of times more than face value i.e. rare coins/outliers

One tool that I just learned on Excel is called a multiple regression analysis that could generate a formula that could describe the approximate MV based on all the variables. Before I attempt this, I'm wondering if anyone has done similar?

kranky

Welcome to the boards, ChuckC.

I think it would be impossible to use any formula that could even remotely calculate the value of a collection based on face value. Some half-cents are worth over a million times face value, and some $20 gold coins are worth 25 times face. Trying to balance all those into a single formula, not knowing what is in a collection, would be impossible.

Perhaps if all the coins were common and of the same type (let's say common date Indian Head cents in VG condition) you could come up with a ballpark number, but if the coins are of various face values and vary widely in market value, no way.

relayer

There is no relation between the face value and the market value.

You can use the other criteria you have listed but it is not consisent across the board.

The value is what a willing buyer will pay a willing seller in an arms length transaction.

Baley

A Related Thread

ChuckC

It's true...one would have to come up with a coefficient for every single coin and variety in every condition by every grader, and the coefficient would have to be updated just about every day. Here's what I mean...

Let's say the only coin you have in your collection is a 1941-D mercury dime graded MS65FB by PCGS, and let's assume the last ten auctions for this particular coin averaged $30. The Regression formula for this exact coin would be:

y = a + bx1 where:
y is the market value
a is the constant ($0.10)
b is the coefficient for a 1941-D mercury dime graded MS65FB by PCGS ($29.90)
x1 is whether or not you have this coin (0 = no, 1 = yes).

Thus the equation for this collection would be y = 0.10 + 29.9(1) which equals $30.00.

Adding a second coin to the collection (e.g. same coin, MS66FB by PCGS) would add another variable to the fomula. Then it would read:
y = a + bx1 + cx2

And so on until all coins, varieties, grades, and graders are accounted for thus making the equation impossibly long. Not to mention other factors involved such as demand or tonign characteristics.

In other words.. F it!

thanks for reading!

Gonfunko

to the forums....I'm not quite sure exactly what you're asking, and I refuse to do math since it's still summer break for a few more days, but William Sheldon proposed a formula for determining the value of large cents based on the grade and vice-versa in his book Penny Whimsy. If you don't have it already, I'd highly reccomend it, especially if you're into early copper.

mr1931S

Coin values are mainly established by demand. Since demand for particular coins are always in a state of flux, you would have to dynamically program your regression analysis to account for demand. You mentioned demand as a variable to consider. That's good but throw out precious metal content and current spot price as these factors are mostly irrelevant for truly collectible coins.

Type, condition, and, to a lesser extent, scarcity are constants but factoring the demand variable correctly into your analysis will be your biggest problem in coming up with anything meaningful.

Good luck, nonetheless.