Comments on: The 75% Problem http://zemblangrammar.com/2008/06/09/the-75-problem/ Alfina the Vague: hermeneutic detective. Wed, 07 Jan 2009 16:53:17 +0000 http://wordpress.org/?v=2.7 hourly 1 By: John http://zemblangrammar.com/2008/06/09/the-75-problem/comment-page-1/#comment-2313 John Wed, 18 Jun 2008 16:58:39 +0000 http://zemblangrammar.com/?p=326#comment-2313 Vague: I'm still pondering this 75% problem you've posed. I've found a quote that expresses what I want to say better than I could say it: <i>... truly to enjoy bodily warmth, some small part of you must be cold, for there is no quality in this world that is not what it is merely by contrast. Nothing exists in itself. If you flatter yourself that you are all over comfortable, and have been so a long time, then you cannot be said to be comfortable any more.</i> So, basically, don't we in a sense actually <i>need</i> the 25%? Isn't it actually less a problem 25% than it is a necessary 25%? Vague:

I’m still pondering this 75% problem you’ve posed. I’ve found a quote that expresses what I want to say better than I could say it:

… truly to enjoy bodily warmth, some small part of you must be cold, for there is no quality in this world that is not what it is merely by contrast. Nothing exists in itself. If you flatter yourself that you are all over comfortable, and have been so a long time, then you cannot be said to be comfortable any more.

So, basically, don’t we in a sense actually need the 25%? Isn’t it actually less a problem 25% than it is a necessary 25%?

]]> By: Timothy http://zemblangrammar.com/2008/06/09/the-75-problem/comment-page-1/#comment-2288 Timothy Sat, 14 Jun 2008 03:08:43 +0000 http://zemblangrammar.com/?p=326#comment-2288 John - Hubris? Because in some cases if you structure your model right you can do better than random guessing? I vaguely remember Markov from back in the day, but it's honestly been four years since I did any real economics. There is a lot of distributed lag in macro, but you know, a lot of economic work is of the "doing blah increases bleh by bloop" variety. John - Hubris? Because in some cases if you structure your model right you can do better than random guessing? I vaguely remember Markov from back in the day, but it’s honestly been four years since I did any real economics. There is a lot of distributed lag in macro, but you know, a lot of economic work is of the “doing blah increases bleh by bloop” variety.

]]> By: John http://zemblangrammar.com/2008/06/09/the-75-problem/comment-page-1/#comment-2286 John Fri, 13 Jun 2008 19:00:11 +0000 http://zemblangrammar.com/?p=326#comment-2286 Timothy: Re: forecasting - I've never understood why economists attempt this. You can't predict random processes unless you want to dispense with the Markov property. Actually even if you do there is still some room to cheat; non-Markovian processes can be cast in a Markovian representation, but that's getting too technical. The best you can do is assume some type of autoregressive behaviour (1 or 2 periods) and try for some confidence intervals, but, you know tomorrow the market can always crash and Markov wins again... I've always favoured the idea that there's an asymmetric behaviour wherein the (recent) past has some influence (e.g. current price level depends on the price at time t-1) but, obviously, the future doesn't. Anyways, not a math blog so I'd best shut up. Timothy:

Re: forecasting - I’ve never understood why economists attempt this. You can’t predict random processes unless you want to dispense with the Markov property. Actually even if you do there is still some room to cheat; non-Markovian processes can be cast in a Markovian representation, but that’s getting too technical. The best you can do is assume some type of autoregressive behaviour
(1 or 2 periods) and try for some confidence intervals, but, you know tomorrow the market can always crash and Markov wins again…

I’ve always favoured the idea that there’s an asymmetric behaviour wherein the (recent) past has some influence (e.g. current price level depends on the price at time t-1) but, obviously, the future doesn’t. Anyways, not a math blog so I’d best shut up.

]]> By: Timothy http://zemblangrammar.com/2008/06/09/the-75-problem/comment-page-1/#comment-2285 Timothy Fri, 13 Jun 2008 18:41:53 +0000 http://zemblangrammar.com/?p=326#comment-2285 V - Well, he does like Jazz so your father is obviously a man of some taste (I lack essentially all Jazz knowledge, but will listen to it gladly whenever it's played), but the Evanescence is really a deal breaker. You've lucked out this time, Vague! St. A - Thinking about it, we have that same thing in economics (except when the model fails we blame the data). Out of sample forcasting is, as we kids say, a bitch. V - Well, he does like Jazz so your father is obviously a man of some taste (I lack essentially all Jazz knowledge, but will listen to it gladly whenever it’s played), but the Evanescence is really a deal breaker. You’ve lucked out this time, Vague!

St. A - Thinking about it, we have that same thing in economics (except when the model fails we blame the data). Out of sample forcasting is, as we kids say, a bitch.

]]> By: David Bowie Circa 1973 http://zemblangrammar.com/2008/06/09/the-75-problem/comment-page-1/#comment-2281 David Bowie Circa 1973 Thu, 12 Jun 2008 22:23:59 +0000 http://zemblangrammar.com/?p=326#comment-2281 I've broken up with women for not being quite as asexual as I would like. And for also not having penises. Perhaps this makes me shallow. I’ve broken up with women for not being quite as asexual as I would like. And for also not having penises. Perhaps this makes me shallow.

]]>