Comments on: Learning to predict the future http://www.vetta.org/2009/02/learning-to-predict-the-future/ Fri, 03 Sep 2010 19:53:44 +0000 hourly 1 http://wordpress.org/?v=3.0.1 By: Shane Legg http://www.vetta.org/2009/02/learning-to-predict-the-future/comment-page-1/#comment-19189 Shane Legg Fri, 20 Feb 2009 23:29:15 +0000 http://www.vetta.org/?p=141#comment-19189 Kelvin Ditto for me. I thought the credit crisis was over around the middle of last year! It had already been in the news for quite a while and things seems to be calming down. Just as well I continued to wait before re-entering the stock market. Picking the true bottom of the market probably isn't that important. My guess is that in 5 years time now will look like a pretty good buying period even if the market goes down a bit more in the next year. Kelvin

Ditto for me. I thought the credit crisis was over around the middle of last year! It had already been in the news for quite a while and things seems to be calming down. Just as well I continued to wait before re-entering the stock market.

Picking the true bottom of the market probably isn’t that important. My guess is that in 5 years time now will look like a pretty good buying period even if the market goes down a bit more in the next year.

]]> By: Kelvin Hartnall http://www.vetta.org/2009/02/learning-to-predict-the-future/comment-page-1/#comment-19188 Kelvin Hartnall Fri, 20 Feb 2009 22:36:41 +0000 http://www.vetta.org/?p=141#comment-19188 As I’ve been watching the share markets plunge and trying to predict when to resume buying, I’ve found myself reflecting on your post. I think the issue is that technology changes faster than society. So we can predict the rate of change of technology reasonably well, though when we use the same rate of change on society we tend to consistently over-estimate how quickly things will change. I suspect that the society’s mood moves slower again, and therefore our mistakes here are even more pronounced. My biggest consistent mistake has been over-estimating how quickly the mood of euphoria or despair will switch. For example, when the NASDAQ reached 3000 in November 1999 I was so adamant that the bubble would burst that I considered buying PUT options. However, the euphoria continued until the NASDAQ reached over 5000! Another example: I felt there was a significant housing bubble in 2004, but the euphoria continued for many more years and house prices increased significantly in this time. I now feel that the markets have reached a bottom, though trying to use an adjustment factor to account for my over-estimation in the past, I assume that the switch from despair to optimism will be a lot further out than I would normally predict. As I’ve been watching the share markets plunge and trying to predict when to resume buying, I’ve found myself reflecting on your post. I think the issue is that technology changes faster than society. So we can predict the rate of change of technology reasonably well, though when we use the same rate of change on society we tend to consistently over-estimate how quickly things will change. I suspect that the society’s mood moves slower again, and therefore our mistakes here are even more pronounced. My biggest consistent mistake has been over-estimating how quickly the mood of euphoria or despair will switch. For example, when the NASDAQ reached 3000 in November 1999 I was so adamant that the bubble would burst that I considered buying PUT options. However, the euphoria continued until the NASDAQ reached over 5000! Another example: I felt there was a significant housing bubble in 2004, but the euphoria continued for many more years and house prices increased significantly in this time. I now feel that the markets have reached a bottom, though trying to use an adjustment factor to account for my over-estimation in the past, I assume that the switch from despair to optimism will be a lot further out than I would normally predict.

]]> By: Shane Legg http://www.vetta.org/2009/02/learning-to-predict-the-future/comment-page-1/#comment-19141 Shane Legg Tue, 10 Feb 2009 01:24:15 +0000 http://www.vetta.org/?p=141#comment-19141 Giovanni: Only if these Christians are wanting to predict the general trends in the stock market... Giovanni:

Only if these Christians are wanting to predict the general trends in the stock market…

]]> By: Shane Legg http://www.vetta.org/2009/02/learning-to-predict-the-future/comment-page-1/#comment-19140 Shane Legg Tue, 10 Feb 2009 01:22:19 +0000 http://www.vetta.org/?p=141#comment-19140 David It's a great company. Buy their product. Tell everybody you know about it. Get you friends to buy one. Did I mention that I'm a (very minor) shareholder? :-) Seriously, I've haven't been on the "inside" of the company for some years and I don't know the IVR market at all. So your guess is as good as mine. One thing I do believe in, however, is Peter's unbelievable tenacity when it comes to getting things to work. That said, I don't think they have the kind of major AGI breakthrough that many a waiting for. David

It’s a great company. Buy their product. Tell everybody you know about it. Get you friends to buy one.

Did I mention that I’m a (very minor) shareholder?

:-)

Seriously, I’ve haven’t been on the “inside” of the company for some years and I don’t know the IVR market at all. So your guess is as good as mine. One thing I do believe in, however, is Peter’s unbelievable tenacity when it comes to getting things to work. That said, I don’t think they have the kind of major AGI breakthrough that many a waiting for.

]]> By: Giovanni http://www.vetta.org/2009/02/learning-to-predict-the-future/comment-page-1/#comment-19137 Giovanni Mon, 09 Feb 2009 18:26:44 +0000 http://www.vetta.org/?p=141#comment-19137 “The evidence so far shows that I’m consistently good at predicting what will happen, but that I predict that it will happen much sooner than it actually does; I roughly need to double my time estimates.” Do you realize this makes you immensely valuable for any decent christian adventist sect? :-) Btw, your point on fads and fashions in technological trends is quite true. Physicists working on social systems use to say that all one is trying (or would like) to do is to model a bunch of electrons with their charge, their position, their speed -- and their opinions. Not very encouraging IMHO ... “The evidence so far shows that I’m consistently good at predicting what will happen, but that I predict that it will happen much sooner than it actually does; I roughly need to double my time estimates.”

Do you realize this makes you immensely valuable for any decent christian adventist sect? :-)

Btw, your point on fads and fashions in technological trends is quite true. Physicists working on social systems use to say that all one is trying (or would like) to do is to model a bunch of electrons with their charge, their position, their speed — and their opinions. Not very encouraging IMHO …

]]> By: David http://www.vetta.org/2009/02/learning-to-predict-the-future/comment-page-1/#comment-19135 David Mon, 09 Feb 2009 12:42:43 +0000 http://www.vetta.org/?p=141#comment-19135 Not entirely related to the post but considering you are well positioned to comment: What are your thoughts on the AGI developments by adaptive ai? They recently launched their first commercial product, an IVR system. Do you think they will be making a big difference in that market and in their commercial applications of AGI research in general? Is this something to be looking out for in terms of significant steps going forward? Not entirely related to the post but considering you are well positioned to comment:

What are your thoughts on the AGI developments by adaptive ai? They recently launched their first commercial product, an IVR system. Do you think they will be making a big difference in that market and in their commercial applications of AGI research in general? Is this something to be looking out for in terms of significant steps going forward?

]]>