"I have x dollars to dispose over y time-span. There are (a, b, c, d, e, f, . . .) products available for purchase. The expected yielded pleasure from each possible selection from the set of possible purchases is as follows: . . ." We can't even answer this question in ordinal terms. This approach is obviously intractable, as these products provide qualitatively different effects on affect, also dependent on interrelation to consumption of other goods, but also psychological state, social context, immediate environment, etc.** That, and nearly no one things thinks out in this way.