Costanza comments on New Year's Predictions Thread (2011) - Less Wrong

10 Post author: Kevin 02 January 2011 09:58AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (224)

You are viewing a single comment's thread.

Comment author: Costanza 02 January 2011 05:00:22PM *  5 points [-]

As noted by Keynes, markets can remain irrational a lot longer than you and I can remain solvent. However, I'd say about 80% for each of the following:

The price of gold will fall in real terms. [ETA one year on -- Wrong. Fell a bit towards the end of the year, but still a net rise.]

The price of U.S. housing will fall in real terms. [ETA one year on -- numbers still seem to be being crunched, or at least I can't find anything very recent I'd regard as reliable.]

The rate of inflation of U.S. dollars will rise. [ETA one year on -- I think I got this one right. Subject to updating with better data in the near future.]

There will be much more discussion in the general media of a "higher education bubble." Not sure how to measure this.

@JoshuaZ, the number of servicemembers leaving due to DADT repeal will not be measured. The goverment won't ask, and won't tell. [ETA one year on -- I think I got this one right, for what it's worth. As I understand it, the U.S. military is on a path toward shrinking in general anyway.]

And finally -- not necessarily a prediction for this year -- things that can't go on forever won't. 100%

Comment author: JoshuaZ 02 January 2011 05:25:36PM 3 points [-]

the number of servicemembers leaving due to DADT repeal will not be measured. The goverment won't ask, and won't tell.

Yes, but others will be trying to measure this, including some aspects of the right-wing in the US.

Comment author: sketerpot 02 January 2011 06:13:35PM *  3 points [-]

I'll go a bit further and predict that, because of how politicized the issue is, there will not be any mainstream agreement on whose unofficial figures are more accurate.

Comment author: Vaniver 02 January 2011 06:32:40PM 2 points [-]

There will be much more discussion in the general media of a "higher education bubble." Not sure how to measure this.

I disagree, for a narrow definition of "general media," and agree for a broad definition of it. It's been an open secret for quite some time that college has a lot of flaws and is negative value for wide swaths of the population, and I imagine that will catch on in many circles. But I really don't see that breaking into the mainstream media, since the value of college is artificially high in most of their narratives (especially since the journalists who are working for the MSM will be people who need to double down on their belief that their degree was worth it).

Comment author: Costanza 02 January 2011 06:57:25PM *  2 points [-]

I should have thought of this before: I just googled "higher education bubble" in quotes, and got 'About 41,000 results'. A search in google news gives five results. A year from now, that should be higher. Not a very good yardstick, but at least it's checkable.

Comment author: Jonathan_Graehl 03 January 2011 01:34:01AM 1 point [-]

You'll need the relative frequency of the phrase (per running word of news article text), not the absolute count of it. In the extreme, if the "from" date of the google news search doesn't advance, you can only expect the number to increase :) Even if there's a fixed time window, you can also expect more words to be generated in the next year than in the past year.

Comment author: Costanza 02 January 2012 05:02:56AM 0 points [-]

For what it's worth, my latest search for the same phrase now results in 'About 1,860,000 results.'

Comment author: gwern 07 January 2011 01:14:48AM 1 point [-]
  1. Gold: http://predictionbook.com/predictions/2100
  2. Housing: http://predictionbook.com/predictions/2101
  3. Inflation: http://predictionbook.com/predictions/2102
  4. Skipping higher-ed bubble and no-measurement because I don't know any good way to operationalize them (Google hits seems like a terrible one, even if we could get replicable data from them about the rate)
Comment author: gwern 28 August 2012 01:17:08AM 0 points [-]
Comment author: gwern 10 January 2012 11:01:58PM 0 points [-]

The rate of inflation of U.S. dollars will rise. [ETA one year on -- I think I got this one right. Subject to updating with better data in the near future.]

Checking CronoDAS's prediction, the core rose to something like 2.15%, which was definitely above what it was last time, so I marked it right.

Comment author: orthonormal 07 January 2012 11:33:43PM 0 points [-]

I'd mention that it's much better to reply to your predictions a year on than to edit them, but then I remembered who I'm talking to.

Comment author: Desrtopa 07 January 2011 02:05:40AM 0 points [-]

@JoshuaZ, the number of servicemembers leaving due to DADT repeal will not be measured. The goverment won't ask, and won't tell.

Was that another prediction, or something the government has already confirmed?

Comment author: Costanza 07 January 2011 03:33:17AM *  2 points [-]

That would be a prediction, specifically applied to what the federal government will do or not do during 2011.

With that said, I'd say human motivations are complicated. I could foresee, for instance the following motivations for leaving the military:

Enlisted infantryman:

[Subjective motivation]: I have failed ranger school in an embarrassing fashion.

[Declared motivation for leaving the army]: A bunch of fags are running the show now.

Infantry officer:

[Subjective motivation]: A bunch of fags are running the show now.

[Declared motivation]: I now wish to serve my country by pursuing public office.

Comment author: orthonormal 07 January 2011 03:48:45AM 0 points [-]

I took it as a joke.

Comment author: PhilGoetz 03 January 2011 11:11:31PM 0 points [-]

Why do you think the price of US housing will fall in real terms (even not counting inflation)?

Comment author: Costanza 03 January 2011 11:42:39PM *  4 points [-]

Basically, I'm persuaded by the arguments that housing prices are out of line with a historical trend and out of line with incomes and prevailing rents.

Now that you mention it, and upon reflection, I must admit I am biased by some vague, aesthetic sentiments as well. My subjective impression is that the bigger=better approach to housing in America should become less fashionable rather than more. I'd like to think of this as an "intuition," but maybe it's just a "bias." Anyway, we'll get some evidence as the new year goes on.

Comment author: orthonormal 04 January 2011 04:39:01AM 1 point [-]

Now that you mention it, and upon reflection, I must admit I am biased by some vague, aesthetic sentiments as well.

Are you going to therefore revise your estimate downward?

Comment author: Costanza 04 January 2011 05:12:31AM *  1 point [-]

That's a good question. If this were a disinterested Bayesian search for the truth, then of course. But I take this particular game as a simulated bet, albeit for no monetary stakes. Before I post anything, I have to search inside myself for the most rational course; but once I place my bet, I'm stuck with it. No fair changing it. Reality will determine whether I win or lose, come the end of 2011.

Also, in this real world of ours, I think inarticulable intuitions may have a rational basis, in the same way that a skilled basketball player can make a free throw without knowing what the hell a parabola is. With that said, I should say I don't yet know whether my predisposition was a "bias" or a "brilliant intuition."

Comment author: orthonormal 04 January 2011 06:08:15AM 2 points [-]

On the other hand, this could make for good, low-stress practice at updating estimates based on new evidence/deductions. You probably should keep your original estimate in its place (since that's what people's replies are predicated on), but could you say here what you'd now estimate that probability to be?

Comment author: Costanza 04 January 2011 06:55:56PM *  1 point [-]

Another good question. Upon further reflection, I'd have to admit that my thoughts in this area are....what's the opposite of rigorous? Fuzzy, vague, unconscious? Someone who is very experienced and skilled in an area that requires intangible intuition or a "feel" for something might have very good judgment without being able to articulate the bases for that judgment. But I'm not experienced or skilled in the housing market. For that reason, I think I should be very suspicious of my own subjective intuitions in this area. They are breeding grounds for unknown bias. So I should revise my estimate downward. I must also admit that any pretense of a specific percentage estimate is far more precise than I am able to give. Right now I'm thinking of a range rather than a point, something like sixty to eighty percent.

P.S. Goddammit, I had to think about this some more and must clarify. My confession: I had to revise specifically downward because I think my subjective emotional biases would tend to make me wish that the housing market came down. On a conscious level, I made my first estimate in good faith. But now I am a bit more wary of what I was then thinking.

Fucking rationality. Okay, I've got stuff to do, so I'm going to forget this topic entirely for a bit.