Peak Oil
Donate Bitcoin

Donate Paypal


PeakOil is You

PeakOil is You

Peak Oil and Bayes' Theorem

General discussions of the systemic, societal and civilisational effects of depletion.
Post a reply

Re: Peak Oil and Bayes' Theorem

Postby WebHubbleTelescope » Tue 27 Dec 2005, 14:17:09

Let's say we want to figure out how fast a particular man walks. Knowing nothing about his ability, initially we say he walks at 1 meters per second. We then measure his speed one day and it comes out 1.3 m/s. According to Bayes, this updates to (1+1.3)/2 = 1.15. If we measure him again at some later date and get 1.3 again it will come to (1.15)*2/3 +1.3/3=1.2.

So the ony real difference between Bayes and some other averaging process is the a priori estimates put on the numbers.

To use Bayes effectivly, you have to find some metric that you have some good prior knowledge about. For the walker, I would have been better off simply measuring how fast the man waks and then kept up a moving average. For those two measurements, I would have gotten 1.3 m/s instead of the Bayesian estimate of 1.2.

Of course there is value in doing all this stuff, as we use this concept every day of our lives in our own personal decision making.

I would suggest instead to apply Bayes theorem to the probability of making new crude oil discoveries. Lots of a prior knowledge built up with new data coming in every year which allows us to update the estimates. I don't know, for some reason, this approach seems a bit more, let me see, practical, than the wankery discussed so far?

And if I have got the particulars of Bayes wrong, well just shoot me, because I am a practical kind of guy.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby EnergySpin » Tue 27 Dec 2005, 14:23:22

$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')

I would suggest instead to apply Bayes theorem to the probability of making new crude oil discoveries. Lots of a prior knowledge built up with new data coming in every year which allows us to update the estimates.

This is an interesting suggestion, but new oil discoveries are based on Bayesian inference tools :-D
"Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
User avatar
EnergySpin
Intermediate Crude
Intermediate Crude
 
Posts: 2248
Joined: Sat 25 Jun 2005, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby WebHubbleTelescope » Tue 27 Dec 2005, 14:35:50

$this->bbcode_second_pass_quote('EnergySpin', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')

I would suggest instead to apply Bayes theorem to the probability of making new crude oil discoveries. Lots of a prior knowledge built up with new data coming in every year which allows us to update the estimates.

This is an interesting suggestion, but new oil discoveries are based on Bayesian inference tools :-D


Which is my point exactly. Let's get some numbers out there so people can see a practical Bayes application in action.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby EnergySpin » Tue 27 Dec 2005, 14:57:50

$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('EnergySpin', '')$this->bbcode_second_pass_quote('WebHubbleTelescope', '
')

I would suggest instead to apply Bayes theorem to the probability of making new crude oil discoveries. Lots of a prior knowledge built up with new data coming in every year which allows us to update the estimates.

This is an interesting suggestion, but new oil discoveries are based on Bayesian inference tools :-D


Which is my point exactly. Let's get some numbers out there so people can see a practical Bayes application in action.

But how is this relevant to the Olduvai Gorge hypothesis?
Oops now I see the problem: there is a disconnect between the thread title (Peak Oil) and the debate within this thread (the plausibility of the Olduvai-back-to-the-stone-age hypothesis).
Rockdok has provided references to the Bayesian machinery as applied to oil exploration, a few months ago. I guess you can ask him if he can dig them up again, but the problem I see with this exercise is the following:
1) lack of data
2) the guys coming up with the priors, are the ones with the MBAs not the ones with the PhDs :P
The combination of (1)+(2) will kill any analysis, and I'm pretty sure that they do not even check the marginal distributions of the discoveries to see they are wrong 8O
"Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
User avatar
EnergySpin
Intermediate Crude
Intermediate Crude
 
Posts: 2248
Joined: Sat 25 Jun 2005, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby entropyfails » Tue 27 Dec 2005, 15:59:11

$this->bbcode_second_pass_quote('EnergySpin', '
')Where did the figure "3" come from?
Is it 3 kwh per person? If it is 3 kwh per person, sorry but the WEC has said that an average of 2.5 (the 4-2-1 policy) is not only attainable but also compatible with advanced technical civilization.


The 3 EUP number comes from the barrels of oil equivalent/capita which OT uses as its metric. Since he uses the 1930's as the start of modern industrial civilization, he sets the end point of industrial civlization when the EUP reaches its 1930's level. The 2030 number comes from that calculation.

$this->bbcode_second_pass_quote('EnergySpin', 'B')y the way , has his article received any citations in the scientific press? Usually articles who make extraordinary claims receive either a lot of attention (when they are worth exploring further or debunking as in cold fusion) or no attention at all (when the blunder is obvious).


I haven’t seen any. I don’t think it has ever had any other form of publication other than the net. If Jay Hanson of dieoff.org didn’t love OT so much, I doubt it would have as large of a following as it does. Dr Duncan hasn’t updated the theory in a couple years. He maintains no website and has no publicly available email address.

As far as I can tell, Dr Duncan has posted a paper as follows.

“The world will blow up in 10 years”
MAGIC er.. SCIENCE HAPPENS HERE
“Kiss your ass goodbye, the end.”

Now with many ways we have of killing ourselves, critical EUP failure does seem to have a certain plausibility given our large population size. People inclined to believe that civilization will kill itself have picked it up and run with it because it has a near term prediction, as opposed to some of the more reasonable longer term predictions of civilization failure. This ends up being doubly true because the paper has some graphs and some math, which tends to make peoples eyes glaze over and turns on the “undoubtedly true” switch. However the OT theory itself seems somewhat testable before the cliff event, and people talk a lot about it, so I thought I would try to raise the bar on the discussion a bit.

I would like to see interest in doing analysis on predictions from the other theories bandied about as well. I like stepping on the magic hobgoblins. Good knowledge always interests me more. *grin*
EntropyFails
"Little prigs and three-quarter madmen may have the conceit that the laws of nature are constantly broken for their sakes." -- Friedrich Nietzsche
User avatar
entropyfails
Expert
Expert
 
Posts: 565
Joined: Wed 30 Jun 2004, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby EnergySpin » Tue 27 Dec 2005, 16:35:14

$this->bbcode_second_pass_quote('entropyfails', '
')
The 3 EUP number comes from the barrels of oil equivalent/capita which OT uses as its metric. Since he uses the 1930's as the start of modern industrial civilization, he sets the end point of industrial civlization when the EUP reaches its 1930's level. The 2030 number comes from that calculation.

So I was right about the FF dependence; sorry entropyfalls, the theory is even more flawed i.e. it contains internal inconsistnecies.Hint: Uranium is a far more abundant and more dense energy source.

$this->bbcode_second_pass_quote('entropyfails', '
')I haven’t seen any. I don’t think it has ever had any other form of publication other than the net. If Jay Hanson of dieoff.org didn’t love OT so much, I doubt it would have as large of a following as it does. Dr Duncan hasn’t updated the theory in a couple years. He maintains no website and has no publicly available email address.

No, he has presented it in a journal, I will look it up for you (this is where I know of this "theory", jerking off on the library e-search service a few years ago). My first impression was: "the reviewers must have been drank, when they read this". But IIRC it was published in one of the "designer journals" i.e. we will publish anything for $ (but I may be wrong on that one).


$this->bbcode_second_pass_quote('entropyfails', '
')Now with many ways we have of killing ourselves, critical EUP failure does seem to have a certain plausibility given our large population size. People inclined to believe that civilization will kill itself have picked it up and run with it because it has a near term prediction, as opposed to some of the more reasonable longer term predictions of civilization failure. This ends up being doubly true because the paper has some graphs and some math, which tends to make peoples eyes glaze over and turns on the “undoubtedly true” switch.

Yeap I know the kind of people that will immediately sign up for such a theory.

$this->bbcode_second_pass_quote('entropyfails', '
') However the OT theory itself seems somewhat testable before the cliff event, and people talk a lot about it, so I thought I would try to raise the bar on the discussion a bit.

Well, the problem with the theory is that would be a pretty good idea of what might happen if all energy sources have been exhausted. There is no peak uranium, peak sun or peak wind from where I stand ... and the ROI of these energy sources to either electiricity or kinetic energy is far greater than petroleum would ever be able to do. Even in the golden petroluem days (the first days), the quoted EROEI of 100 is somewhat misleading. What was the efficiency of oil utilization back then?

$this->bbcode_second_pass_quote('entropyfails', '
')I would like to see interest in doing analysis on predictions from the other theories bandied about as well. I like stepping on the magic hobgoblins. Good knowledge always interests me more. *grin*
Good for you :-D
"Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
User avatar
EnergySpin
Intermediate Crude
Intermediate Crude
 
Posts: 2248
Joined: Sat 25 Jun 2005, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby WebHubbleTelescope » Tue 27 Dec 2005, 18:05:09

$this->bbcode_second_pass_quote('EnergySpin', '
')But how is this relevant to the Olduvai Gorge hypothesis?


I wasted a bit of my time trying out one of Duncan's systems models a few days ago. He used a commercial tool called Stella, which has a trial version that I loaded his world production forecasting model in to.
http://mobjectivist.blogspot.com/2005/1 ... model.html

If anyone sees anything of any significance beyond a basic bean-counting exercise in Duncan's analysis, I would like to know.
User avatar
WebHubbleTelescope
Tar Sands
Tar Sands
 
Posts: 950
Joined: Thu 08 Jul 2004, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby EnergySpin » Wed 28 Dec 2005, 04:40:45

$this->bbcode_second_pass_quote('WebHubbleTelescope', '')$this->bbcode_second_pass_quote('EnergySpin', '
')But how is this relevant to the Olduvai Gorge hypothesis?


I wasted a bit of my time trying out one of Duncan's systems models a few days ago. He used a commercial tool called Stella, which has a trial version that I loaded his world production forecasting model in to.
http://mobjectivist.blogspot.com/2005/1 ... model.html

If anyone sees anything of any significance beyond a basic bean-counting exercise in Duncan's analysis, I would like to know.

The only significance I see in Duncan's theory has to do with the exponentially increasing clinic visits/curb-site questions I get regarding the best anti-depressant. And I need no Bayesian tools, to find out they are related to PO :roll:.
If I were a prominent doomer (of the die-off, will rape and eat your grandma variety), I would sincerely hope that TSHTF or else ...
(angry people who bought into the Apocalypse, will find a good use for the weapons/ammo they hoarded during the preparation phase).
Last edited by EnergySpin on Thu 29 Dec 2005, 08:54:58, edited 1 time in total.
"Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
User avatar
EnergySpin
Intermediate Crude
Intermediate Crude
 
Posts: 2248
Joined: Sat 25 Jun 2005, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby entropyfails » Thu 29 Dec 2005, 07:46:51

$this->bbcode_second_pass_quote('EnergySpin', '')$this->bbcode_second_pass_quote('entropyfails', '
')
The 3 EUP number comes from the barrels of oil equivalent/capita which OT uses as its metric. Since he uses the 1930's as the start of modern industrial civilization, he sets the end point of industrial civlization when the EUP reaches its 1930's level. The 2030 number comes from that calculation.

So I was right about the FF dependence; sorry entropyfalls, the theory is even more flawed i.e. it contains internal inconsistnecies.Hint: Uranium is a far more abundant and more dense energy source.

I still don’t see this criticism as valid. The OT measures things in Joules per capita, by way of the number of Joules in a barrel of oil and dividing by the total energy use by that number. He just uses a static conversion number, (I don’t remember exactly what). OT takes into account all of the direct energy use of the planet by humans and seems agnostic to the energy type in theory, if not in how people portray it.

Honestly, he would have done better by simply listing what he felt was the minimum Joule per capita requirement for civilization. The reason for his choosing barrels of oil equivalent probably comes from the want to relate the information to the laymen. Perhaps a secondary reason would be to create an unconscious link between the well understood decline in oil with his supposed decline in EUP. As you point out, that doesn’t have to be the case. I feel certain he would counter that once we have peaked in EUP, no other technology can replace the energy use faster than the population that grows to use it, hence EUP will always drop after the peak.

Of course, determining the validity of that statement ends up being my plan. *grin*
Last edited by entropyfails on Fri 30 Dec 2005, 07:01:56, edited 1 time in total.
EntropyFails
"Little prigs and three-quarter madmen may have the conceit that the laws of nature are constantly broken for their sakes." -- Friedrich Nietzsche
User avatar
entropyfails
Expert
Expert
 
Posts: 565
Joined: Wed 30 Jun 2004, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby EnergySpin » Thu 29 Dec 2005, 09:01:03

$this->bbcode_second_pass_quote('entropyfails', '
')Honestly, he would have done better by simply listing what he felt was the minimum Joule per capita requirement for civilization.

Maybe because such a measure does nto exist?

$this->bbcode_second_pass_quote('entropyfails', '
')The reason for his choosing barrels of oil equivalent probably comes from the want to relate the information to the laymen. Perhaps a secondary reason would be to create an unconscious link between the well understood decline in oil with his supposed decline in EUP. As you point out, that doesn’t have to be the case. I feel certain he would counter that once we have peaked in EUP, no other technology can replace the energy use faster than the population that grows to use it, hence EUP will always drop after the peak.

We hit the "peak" in the 70s ... note the time series he uses, have been used by others to pinpoint the growth in energy utilization efficiency as a result of the oil shocks. By the way, civilization (and not just population) CANNOT increase at a rate faster than the energy production x efficiency of use. This part of the theory goes againts the 1st law :wink:

$this->bbcode_second_pass_quote('entropyfails', '
')Of course, determining the validity of that statement ends up being my plan. *grin*

Go for it! At the very least you will get your hands dirty with Bayesian methods.
"Nuclear power has long been to the Left what embryonic-stem-cell research is to the Right--irredeemably wrong and a signifier of moral weakness."Esquire Magazine,12/05
The genetic code is commaless and so are my posts.
User avatar
EnergySpin
Intermediate Crude
Intermediate Crude
 
Posts: 2248
Joined: Sat 25 Jun 2005, 03:00:00
Top

Re: Peak Oil and Bayes' Theorem

Postby entropyfails » Fri 30 Dec 2005, 06:47:29

$this->bbcode_second_pass_quote('EnergySpin', '')$this->bbcode_second_pass_quote('entropyfails', '
')Honestly, he would have done better by simply listing what he felt was the minimum Joule per capita requirement for civilization.

Maybe because such a measure does nto exist?


*grin* As Peter Griffin says "Perhaps..."

$this->bbcode_second_pass_quote('EnergySpin', 'W')e hit the "peak" in the 70s ... note the time series he uses, have been used by others to pinpoint the growth in energy utilization efficiency as a result of the oil shocks. By the way, civilization (and not just population) CANNOT increase at a rate faster than the energy production x efficiency of use. This part of the theory goes againts the 1st law :wink:

WARNING! YOU HAVE APPLIED THE LAWS OF THERMODYNAMICS TO A NON CLOSED SYSTEM!!!

*laugh*

EUP doesn't take into account solar energy use appropriated for humans. EUP can decline in the face of a growing population.
$this->bbcode_second_pass_quote('EnergySpin', '')$this->bbcode_second_pass_quote('entropyfails', '
')Of course, determining the validity of that statement ends up being my plan. *grin*

Go for it! At the very least you will get your hands dirty with Bayesian methods.

*grin* My thoughts exactly! Too bad this thread didn't get more interest from others around here. *sigh*
EntropyFails
"Little prigs and three-quarter madmen may have the conceit that the laws of nature are constantly broken for their sakes." -- Friedrich Nietzsche
User avatar
entropyfails
Expert
Expert
 
Posts: 565
Joined: Wed 30 Jun 2004, 03:00:00
Top
Previous
Post a reply

Return to Peak Oil Discussion

Who is online

Users browsing this forum: No registered users and 1 guest



cron