3 ms·

Depends on your model of things. You could bake other assumptions into a distribution explaining why someone like him who has taken these past positions is in l

Depends on your model of things. You could bake other assumptions into a distribution explaining why someone like him who has taken these past positions is in less likely than someone else to be right. You assume you we can't say anything informative about his prediction. Disagree.

pizza 9y ago
> You assume you we can't say anything informative about his prediction. OK, can you show me a model that is better? What I'm saying is not dependent on the arbitrariness of subjective "baked-in assumptions", nor that each market observer has a subjective perspective and is not omniscient, nor is it about whether or not you may have the gut feeling that he's trying to scam you into unnecessary doomsday preparation. In stats terms: not about how each of us have different perspectives so KL-divergence > 0. What I am saying is: all of the people who say "because these past positions never definitely happened yet, we should be more skeptical of his predictions than everyone else saying things will be just fine" are getting it wrong. You should be extremely skeptical of everyone asking you to withhold skepticism as if past observations (samples) prove the distribution is a stationary ergodic process (truly observing the generator). When someone has predicted things successfully in the past, it means they answered a falsifiable question before it - so there is a burden of proof before you can call them a good predictor. But a prediction that has never been falsified does not demand a burden of proof, you can't say it's incorrect or correct at all - no matter the duration, the question "are his positions wrong?" is meaningless. I am not privileging his opinion that much, tbh - imo, someone saying "catastrophes in the future will be worse" is a nearly-zero surprisal/maximum entropy statement. What influence could I exert over the outcomes that I depend on of processes far beyond what I can control? None. "just the way she goes, boys." All I can do is say memento mori, insure myself against catastrophe (when the cost is negligible, so the waste is low if the catastrophe never came), and not dull my senses.. meaning that it is preferable to ignore salesmen reaffirming one another saying "but we've never sold this much before!" over to ignore someone without special insight who you suspect is only fomenting fear irrespective of if they stand to benefit, deceitfully or not, from that fear. "We can infer Jim Rogers is unlikely to be correct about the future of catastrophes because the last 6 years of recorded performance did not contain a catastrophic event" != "somebody who had said everything is fine 6 years in a row would have been more correct than Jim Rogers over the last 6 years THEREFORE his predictions are less likely to be right", which ignores the point that ).
- petulla 9y ago
  Nope.

pizza 9y ago

> You assume you we can't say anything informative about his prediction. OK, can you show me a model that is better? What I'm saying is not dependent on the arbitrariness of subjective "baked-in assumptions", nor that each market observer has a subjective perspective and is not omniscient, nor is it about whether or not you may have the gut feeling that he's trying to scam you into unnecessary doomsday preparation. In stats terms: not about how each of us have different perspectives so KL-divergence > 0. What I am saying is: all of the people who say "because these past positions never definitely happened yet, we should be more skeptical of his predictions than everyone else saying things will be just fine" are getting it wrong. You should be extremely skeptical of everyone asking you to withhold skepticism as if past observations (samples) prove the distribution is a stationary ergodic process (truly observing the generator). When someone has predicted things successfully in the past, it means they answered a falsifiable question before it - so there is a burden of proof before you can call them a good predictor. But a prediction that has never been falsified does not demand a burden of proof, you can't say it's incorrect or correct at all - no matter the duration, the question "are his positions wrong?" is meaningless. I am not privileging his opinion that much, tbh - imo, someone saying "catastrophes in the future will be worse" is a nearly-zero surprisal/maximum entropy statement. What influence could I exert over the outcomes that I depend on of processes far beyond what I can control? None. "just the way she goes, boys." All I can do is say memento mori, insure myself against catastrophe (when the cost is negligible, so the waste is low if the catastrophe never came), and not dull my senses.. meaning that it is preferable to ignore salesmen reaffirming one another saying "but we've never sold this much before!" over to ignore someone without special insight who you suspect is only fomenting fear irrespective of if they stand to benefit, deceitfully or not, from that fear. "We can infer Jim Rogers is unlikely to be correct about the future of catastrophes because the last 6 years of recorded performance did not contain a catastrophic event" != "somebody who had said everything is fine 6 years in a row would have been more correct than Jim Rogers over the last 6 years THEREFORE his predictions are less likely to be right", which ignores the point that ).

petulla 9y ago
Nope.