Originally posted by GeorgeK
Sportacle's approach is sound. In all these basic models, the principles are very solid --- it's a matter of how you put in the inputs, though, as there are assumptions in each model. E.g. one of the Discounted Cash Flow model's inputs is the growth rate. That's usually not constant over time, either (there are models that allow for two growth rates, e.g. one in the short-run, "hypergrowth", and then a long-run growth rate which is lower).
Even assets with negative cash flows can be valued using various approaches. For example, suppose you have a gold mine, where the cost of production is $500/ounce. Yet, the current price of gold is only $400/ounce. Is the gold mine worthless? No -- you have an embedded option, in that the gold mine will payoff if gold rises. So, like a stock option, that "out of the money" option has value....and it has value the longer the term of the option.
e.g. a call option with an exercise price of $100, when the stock is currently at $50, is practically worthless if there's only 1 week until expiry of the option (unless the volatility is huge!). But, if that option is for 10 years, the option will have considerable value.
Similarly with domain values....in a way, you can look at some of them as "options"....they might pay off big one day, and the cost of keeping the option "alive" is the continued registration cost of $6 to $10/yr. If you believe domain values in the future are very volatile, that 'option' will have value.