:) It was just a good month—most of the entries are correlated: the fall of SR set off the spectacular rise in Bitcoin, the bet was based on unwarranted pessimism re the rise, the blackmarket turmoil made my expertise in blackmarkets of unusual interest to mainstream media, the turmoil prompted me to try to puncture unwarranted optimism by doing the public bet and also the survival analysis & DPR estimation, the Bitcoin rise also helped prompt the Sheep theft (which then unlocked the doxing)...
The only really uncorrelated parts are finally finishing Radiance (which was just a matter of time), and being contacted out of the blue with 2 new Zeo datasets.
I follow you on G+. It’s the unrelenting, what-I-did-this-month posts, every month, that I find so impressive.
Btw, I had a random thought the other day. I don’t remember reading your thoughts on GiveWell, but I see that in the past you’ve done freelance research and writing for MIRI and others, and it seems to me that your brand of efficient research (esp. w/ your emphasis on statistical analysis) might be a good fit for some of GW’s shallow investigations.
Not sure if you’re looking for more work beyond all the projects you’ve already got going on, but I just wanted to pass that idea along, in case it might seem promising and hadn’t occurred to you.
I follow you on G+. It’s the unrelenting, what-I-did-this-month posts, every month, that I find so impressive.
Oh. Fair enough! I started the monthly posts as a trial run for seeing whether I could handle a monthly summary or not, before I tried starting up a mailing list. It seemed to go pretty well so I decided to open one up this December.
Btw, I had a random thought the other day. I don’t remember reading your thoughts on GiveWell
I thought about it in the past, and was actually why I volunteered for them shortly. I never did write that up, did I? In any case, at the time I had no particular credentials for them or real statistical skills so it didn’t go anywhere. Part of the problem was that they put an emphasis on their employees being physically present at their Chinatown, IIRC, offices in NYC. At the time I was not very near NYC, and I’m even further away now. And considering the idea at the present, Givewell seems to be moving away from statistical approaches. So… It’s not a bad idea, but events seem to be conspiring to render it ever more remote an outcome.
At this point, I’m more concerned with whether the doxing was correct. I regard it as a waste of time to write more on the topic until more information (like arrests) come in.
Request for information—should I take this to mean you sold them for dollars?
(I’m asking because I have some btc which I had forgotten about until recently and am wondering if this is the correct time to trade for dollar, but I haven’t done any research on the topic and wouldn’t really know where to start.)
Yup. I had termed it status quo bias, but endowment effect would also explain it.
I was talking it over with a friend, who suggested that what’s really going on is that humans are more willing to gamble with fortuitous winnings than they are with purposeful investments (as in, you’ll be more likely to gamble a dollar you found on the ground than one you earned)… and I actually think that this is the biggest contributor to the explanation.
Of course, holding liquid assets in dollars wasn’t a deliberate decision either—it was just used by default. Even if I was completely risk averse, It’s not like I’ve got evidence that dollars are the most safe asset.
If I was planning this deliberately—I’m still a dependent and don’t really earn yet, so I haven’t really researched this—I’d probably store liquid assets in an index fund or something.
The thing is, when I look at the big picture everything is so riddled with bias already. For example, when Bitcoin began I put a high certainty on the value increasing, but I didn’t deliberately buy any. Before the recession, I put a high certainty on there being a recession before they put the temporary prohibition on shorting, but I didn’t short anything. Yes, it’s a good heuristic to not do anything (or more accurately, follow the majority) in scenarios when you are uncertain of your own competence...but my reason for not doing anything wasn’t explicitly negotiated via that reasoning, it was simply an instinctive tendency to not activity do things. (Interestingly, in this scenario uncertainty about my competence doesn’t imply inaction but action, since I’d have to deliberately buy USD.)
Edit: I’m reminded of this post… I think investing money in small amounts qualifies for “aliveness in training”, although as the amounts get larger it’s not training anymore. Perhaps we should encourage users who wish to practice in rationality to invest some liquid assets in something other than their native currency (unless they’ve deliberately chosen the native currency as the best place to store assets) to practice decision making?
So if you answer no to that question, then you are biased. But it’s not actually obvious which direction. What if non-Bitcoin-owning you is wrong, and you really should be buying it?
Also, if you have a lot of it, the extra Bitcoin might change your marginal value of money (because if you count the value of the BTC, you have more dollars than you otherwise would).
Yeah, but similarly, if you have 2 contradictory beliefs A & B, you don’t know whether to reject A or reject B (or maybe reject the argument which claims their contradictory). It’s like the saying, one man’s modus ponens is another man’s modus tollens. What Miller’s heuristic is useful for is revealing that you need to think and calculate more carefully on this topic.
Also, if you have a lot of it, the extra Bitcoin might change your marginal value of money
Yes, as Bitcoin appreciates, it would probably make more sense to rebalance your portfolio by selling off some bitcoins.
(Personally, I’m risk-seeking at this point, and I think most of the world still doesn’t appreciate a lot of the value in Bitcoin. For example, the blackmarkets still have not universally incorporated basic Bitcoin functionality like multi-sig escrow, and they have huge incentives to do so! If the Bitcoin blackmarkets are so primitive, the rest of the world still has a long ways to go to catch up and buy into Bitcoin. And indeed, I just finished buying some more bitcoins, although at this point I’m starting to get uncomfortable holding so much, so I think I’m stopping where I am.)
Correct, and I was implicitly ignoring income effects. Still, putting a lot of your assets in Bitcoins or any other exotic investment vehicle is ex-ante a HORRIBLE idea unless you have some special informational advantage.
I’d insert “think you should” before “buy”—there’s a large chance that if I didn’t have any bitcoins at the moment I would be keeping on putting off buying some because of the inconvenience.
How risk-averse are you? But even if you aren’t, I suspect that right now bitcoins aren’t a great investment strictly in expected-value terms due to the high risk that they will decline in value by a lot. No one really knows what will happen, though.
I’m fairly risk averse in the sense that I wouldn’t ordinarily feel confident enough in my knowledge to play on speculative markets...I just happen to possess BTC because someone liked something I wrote and kindly tipped me in BTC a couple years ago. I’d forgotten about it for a while, and today realized that it’s worth 20x more than it was originally.
So basically my inaction caused me to inadvertently made a good investment. And even though I wouldn’t ordinarily have tried investing in BTC, now that the decision to invest has essentially been made for me I curiously haven’t immediately decided to convert to dollars.
Status quo bias is interesting—I’m tempted just to keep it and watch what happens...so far it’s still rising. I guess the amount on the line isn’t actually large enough for loss aversion fear to kick in .
Won a big bet, made more money off Bitcoin than I have ever earned normally, bet the world like a boss, did some neat statistics, finished transcribing a great novel, interviewed with Mike Power & the BBC, doxed a drug lord.
Be honest Gwern, are you an upload?
:) It was just a good month—most of the entries are correlated: the fall of SR set off the spectacular rise in Bitcoin, the bet was based on unwarranted pessimism re the rise, the blackmarket turmoil made my expertise in blackmarkets of unusual interest to mainstream media, the turmoil prompted me to try to puncture unwarranted optimism by doing the public bet and also the survival analysis & DPR estimation, the Bitcoin rise also helped prompt the Sheep theft (which then unlocked the doxing)...
The only really uncorrelated parts are finally finishing Radiance (which was just a matter of time), and being contacted out of the blue with 2 new Zeo datasets.
I follow you on G+. It’s the unrelenting, what-I-did-this-month posts, every month, that I find so impressive.
Btw, I had a random thought the other day. I don’t remember reading your thoughts on GiveWell, but I see that in the past you’ve done freelance research and writing for MIRI and others, and it seems to me that your brand of efficient research (esp. w/ your emphasis on statistical analysis) might be a good fit for some of GW’s shallow investigations.
Not sure if you’re looking for more work beyond all the projects you’ve already got going on, but I just wanted to pass that idea along, in case it might seem promising and hadn’t occurred to you.
Oh. Fair enough! I started the monthly posts as a trial run for seeing whether I could handle a monthly summary or not, before I tried starting up a mailing list. It seemed to go pretty well so I decided to open one up this December.
I thought about it in the past, and was actually why I volunteered for them shortly. I never did write that up, did I? In any case, at the time I had no particular credentials for them or real statistical skills so it didn’t go anywhere. Part of the problem was that they put an emphasis on their employees being physically present at their Chinatown, IIRC, offices in NYC. At the time I was not very near NYC, and I’m even further away now. And considering the idea at the present, Givewell seems to be moving away from statistical approaches. So… It’s not a bad idea, but events seem to be conspiring to render it ever more remote an outcome.
Minor: GiveWell is in SF now.
Oh. That makes things even worse then. At least then and now I was on the right coast.
Please consider writing a comprehensive post on this.
Isn’t that one of those “may be hazardous to your health” things?
Given that he already admitted doing it, the marginal harm of providing more detail might not be that high.
At this point, I’m more concerned with whether the doxing was correct. I regard it as a waste of time to write more on the topic until more information (like arrests) come in.
By the way, what you said about the Silk Road bust publicising that something like Silk Road was even possible? It’s made it so far into mainstream awareness that it’s a matter for humour on a completely mainstream parody news site.
We’ll know blackmarkets have really arrived when they show up on The Onion.
Request for information—should I take this to mean you sold them for dollars?
(I’m asking because I have some btc which I had forgotten about until recently and am wondering if this is the correct time to trade for dollar, but I haven’t done any research on the topic and wouldn’t really know where to start.)
If you owned zero Bitcoins would you buy some to speculate? If no then sell, otherwise you are falling victim to the endowment effect bias.
Yup. I had termed it status quo bias, but endowment effect would also explain it.
I was talking it over with a friend, who suggested that what’s really going on is that humans are more willing to gamble with fortuitous winnings than they are with purposeful investments (as in, you’ll be more likely to gamble a dollar you found on the ground than one you earned)… and I actually think that this is the biggest contributor to the explanation.
Of course, holding liquid assets in dollars wasn’t a deliberate decision either—it was just used by default. Even if I was completely risk averse, It’s not like I’ve got evidence that dollars are the most safe asset.
If I was planning this deliberately—I’m still a dependent and don’t really earn yet, so I haven’t really researched this—I’d probably store liquid assets in an index fund or something.
The thing is, when I look at the big picture everything is so riddled with bias already. For example, when Bitcoin began I put a high certainty on the value increasing, but I didn’t deliberately buy any. Before the recession, I put a high certainty on there being a recession before they put the temporary prohibition on shorting, but I didn’t short anything. Yes, it’s a good heuristic to not do anything (or more accurately, follow the majority) in scenarios when you are uncertain of your own competence...but my reason for not doing anything wasn’t explicitly negotiated via that reasoning, it was simply an instinctive tendency to not activity do things. (Interestingly, in this scenario uncertainty about my competence doesn’t imply inaction but action, since I’d have to deliberately buy USD.)
Edit: I’m reminded of this post… I think investing money in small amounts qualifies for “aliveness in training”, although as the amounts get larger it’s not training anymore. Perhaps we should encourage users who wish to practice in rationality to invest some liquid assets in something other than their native currency (unless they’ve deliberately chosen the native currency as the best place to store assets) to practice decision making?
So if you answer no to that question, then you are biased. But it’s not actually obvious which direction. What if non-Bitcoin-owning you is wrong, and you really should be buying it?
Also, if you have a lot of it, the extra Bitcoin might change your marginal value of money (because if you count the value of the BTC, you have more dollars than you otherwise would).
Yeah, but similarly, if you have 2 contradictory beliefs A & B, you don’t know whether to reject A or reject B (or maybe reject the argument which claims their contradictory). It’s like the saying, one man’s modus ponens is another man’s modus tollens. What Miller’s heuristic is useful for is revealing that you need to think and calculate more carefully on this topic.
Yes, as Bitcoin appreciates, it would probably make more sense to rebalance your portfolio by selling off some bitcoins.
(Personally, I’m risk-seeking at this point, and I think most of the world still doesn’t appreciate a lot of the value in Bitcoin. For example, the blackmarkets still have not universally incorporated basic Bitcoin functionality like multi-sig escrow, and they have huge incentives to do so! If the Bitcoin blackmarkets are so primitive, the rest of the world still has a long ways to go to catch up and buy into Bitcoin. And indeed, I just finished buying some more bitcoins, although at this point I’m starting to get uncomfortable holding so much, so I think I’m stopping where I am.)
While that may be true, the major downside risk is the governments putting their foot down. I treat that as “when”, not “if”.
Correct, and I was implicitly ignoring income effects. Still, putting a lot of your assets in Bitcoins or any other exotic investment vehicle is ex-ante a HORRIBLE idea unless you have some special informational advantage.
I’d insert “think you should” before “buy”—there’s a large chance that if I didn’t have any bitcoins at the moment I would be keeping on putting off buying some because of the inconvenience.
No, it’s paper profits so far.
How risk-averse are you? But even if you aren’t, I suspect that right now bitcoins aren’t a great investment strictly in expected-value terms due to the high risk that they will decline in value by a lot. No one really knows what will happen, though.
I’m fairly risk averse in the sense that I wouldn’t ordinarily feel confident enough in my knowledge to play on speculative markets...I just happen to possess BTC because someone liked something I wrote and kindly tipped me in BTC a couple years ago. I’d forgotten about it for a while, and today realized that it’s worth 20x more than it was originally.
So basically my inaction caused me to inadvertently made a good investment. And even though I wouldn’t ordinarily have tried investing in BTC, now that the decision to invest has essentially been made for me I curiously haven’t immediately decided to convert to dollars.
Status quo bias is interesting—I’m tempted just to keep it and watch what happens...so far it’s still rising. I guess the amount on the line isn’t actually large enough for loss aversion fear to kick in .