= 27d567bc2d48d9f40e9ccc20ea370b15)
satt comments on Open thread, Aug. 17 - Aug. 23, 2015 - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (106)
A side note, but my immediate reaction to reading that was "that's overconfidence", though it's quite possible I'm interpreting "financial crash" more grandly than you intended. If I define it — just to try to be more concrete, albeit still pretty fuzzy — as an event as bad as the 2007-8 financial crisis, I'd put a probability ≈5% on that happening by 2015's end.
(Adding the conjunction that explicitly nationalist and/or communist parties gain power actually wouldn't reduce my probability that much, because that already happens; see the Scottish Nationalist Party winning 56 of 59 Scottish seats in this year's general election, and AKEL's performance in Cyprus's 2006 & 2008 elections.)
It may be worth noting that the SNP is quite unlike the far-right nationalist parties I think skeptical-lurker is worried about. They're a party of the left rather than of the right, they aren't outrageously far to the left (at least by European standards), and their "nationalism" is all about getting out of the UK rather than about keeping out immigrants, throwing out black people, and other such traditional policies of nationalist parties.
Indeed. (And my impression is that Demetris Christofias wasn't much of a stereotypical communist as president of Cyprus, either; he seems to've concentrated on the dispute with Turkey.) I was trying to also illustrate another knotty aspect of rigorously operationalizing a broad political prediction like this: which precise kinds of nationalist/communist party gaining power would count as fulfilling the prediction?
I'm basically claiming that I know better than the market, so I'm surprised you're the first person to call me on it. Rigorous definitions are complicated by the fact that the 2007-8 financial crisis took about a year to bottom out - even if the next crash started right now, it (probably) wouldn't bottom out by the end of the year. I'd say around 50% prob of a crash of similar impact as 2008 starting by the end of the year. Perhaps I should get a predictionbook account and register my unusual predictions.
Yes, predictions about nationalism/communism are simple extrapolations. I wish to emphasise that I don't think this is necessary a disaster - the SNP are not your stereotypical nationalists, and are not a huge source of concern IMO, and Japan is an ethno-nationalist state which seems to be functioning perfectly fine.
I do find it surprising/worrying that the frontrunner for the leader of the opposition in the UK is a socialist talking about encouraging growth by printing money, leaving NATO, funding homoeopathy, legislating against toy soldiers, etc., and is mostly rejected by his own party for being too left-wing.
To be fair, I didn't look at the markets either. I asked my gut for a reference class forecast ("How often might I expect this kind of crisis to come along in general...?") while trying to remember my impressions of what economic commentators have said about potential bubbles & such. (I used this sort of tactic to guess rough, first-cut probability estimates in the Good Judgment Project, where I did OK.)
Yeah, that was part of the reason my probability was as low as it was. Actually, reflecting on it a tiny bit more rigorously, I'll nudge it down further to maybe 3%, because I originally rounded "before the end of the year" to "half a year" in my head, but the rest of 2015 is more like a third of a year.
With a looser prediction window like "beginning by the end of 2015, but hitting bottom before the end of 2016", I'd raise my "maybe 3%" to "maybe 12%".
I had to Google that one!
Edit to add: regardless of the probability disagreement, I did upvote you for elaborating.
The financial markets allow you to bet on such things. If you actually believe that 50% probability, you can place a bet in the markets at what would look to you like VERY favourable odds.
Would this involve shorting stocks? That seems dodgy, given that the default is for stock to rise. And if my probability distribution is about 50% crash, 40% surge upwards, 10% stability, then shorting isn't a particularly good idea.
Or maybe its time to buy some bitcoin...
You can express quite complicated views in financial markets.
That distribution doesn't look like "the default is for stock to rise" :-) Besides, a 2008-magnitude crash would probably drive the equity markets to what, half of their current value? You can't expect a "surge upward" to give you symmetrical gains on the upside, so your expected distribution is very lopsided.
Let's make things even simpler -- you believe that in half a year we'll end up in one of two states -- either, with a 50% probability, the equity markets will crash to half price, or, with a 50% probability, the markets will stay the same or go up. If this is so, you can take a position in the markets which is guaranteed to make you money, either a lot (in the first case) or a little (in the second case). Guaranteed, that is, if the markets will do one of these two things.
Its not - stocks rising is the prior, this is my posterior.
Exactly what instruments would guarantee this return? A mixture of shorts and options?
The simplest would be a put spread. You would buy deep out of the money puts and write (sell) other OOM puts much closer to the current price. You will buy more deep OOM puts than sell shallow OOM puts, however because you will be buying cheap puts and selling expensive ones you will start with a positive balance. In the case of markets going nowhere or up, both sets of puts expire worthless and you keep your difference in premium (you earned a little money). In the case of the crash, both sets of puts expire in the money, but you are long more puts than you are short, so you make a lot of money.