[OilyMon]

I've searched for and cannot find, a model describing the price fluctuations of a barrel of oil in terms of the demand. I'd like to go throught the calculations myself and gain an understanding of peak oil from a more scientific and mathematical point of view, in order to supplement my conceptual and semantic understanding. Is there 1 universally accepted model, for the cost per barrel as a function of relative demand? I challenge the forum to prove that supply shortfalls and increasing demand will have the impact that is generally accepted, using mathematical arguments.

Edited for punctuation and clarity.

[tdrive]

...a model describing the price fluctuations of a barrel of oil in terms of the demand...

... does not exist in the public domain. Don't waste your time. There are
proprietary models however, which take a large number of
inputs, and the demand has only a small contribution to the
final price. These, however, require a level of expertise to be
even used and results understood which, based on your post,
you do not have.

You can, however, do a homegrown model based on a historical
multi-dimentional regression on apparent demand, stocks, capacity,
seasonality, price, and derivatives apparent demand, stocks,
capacity, seasonality and prices, for educational purposes only.
Just dont try to trade on this, you will get wiped out faster than
you can say generalized autoregressive conditional heteroskedasticity.

Sorry to burst your bubble. Noone here and anywhere else
will do anything so complicated and expensive just to take your pepsi challenge.

If you have (a lot of) money, however, let me know, I can help you.

Cheers,

[OilyMon]

Thanks for the reply,

How much? What kind of expertise? Where do I buy a model?

[katkinkate]

The price is affected by market sentiment as well, isn't it. The perceptions of the traders and the decisions they make affect price to some extent and no-one has successfully made any mathematic model of emotion, confidence and greed.

[tdrive]

How much? What kind of expertise? Where do I buy a model?

Mister, are you telling me you have a 100k to spare?

If yes, send me a private message with your contact info.

Cheers,

[pup55]

OECD pricing model

lorenzo posted this awhile back. It is reasonably well thought out, and the equations are given in the back. It might save you a few bucks (sorry, tdrive)

There are two problems with it: First of all, they constructed their equations on relatively short-term data (10-year), and tried to do long-term projections, which is always problematic. Secondly, their projections were based on really optimistic (low) demand and economic growth assumptions, that is, they assumed china would grow at 5% long-term and the US at maybe just over population growth rate.

That's why their price projection appears to be comically low.

They do take into account the fact pricing is mainly determined by what minimum amount OPEC decides it should be.

Anyway it's a start.

[khebab]

There is also the short energy model used by the IEA:

Introduction to the Short Term Forecasting Model for petroleum products, natural gas, electricity and utility fuels

[tdrive]

It might save you a few bucks (sorry, tdrive)

Oh, yes, I do not dispute that. There are plenty of academic and
research articles and work done in the field. You can put this
together at home with Excel, use public domain data and will
give you some idea of where things are going, given you can interpret
the results, as I already mentioned.

There is also the short energy model used by the IEA

Unfortunately, STEO takes the oil prices as exogenous variables.
The good thing about it is the query interface: http://tonto.eia.doe.gov/STEO_Query/app/

You may also want to check this: 'World Oil Production Capacity' (WOCAP) model

What I meant was a working model you can run scenarios,
with data interfaces, case study capabilities, what-if scenarios,
tied up to an econometric model, and your corporate trading platform,
tailored to your company objectives. Those run between 100 to 500k
depending on the type of studies you do and the agreed
accuracy, and they are designed to make you money,
not just spit out a number for educational purposes.

As an extreme case there was a project which cost a major energy
company down in Houston about 3M. That thing had a neural network,
energy and weather model of US, live data beamed from sattelite
data collection agency, etc. Pretty cool stuff.

Cheers

[pup55]

Cool stuff is right!

If you have a lot of money at stake, you can afford to go to great lengths to develop your model. Hire some PhD mathematicians and buy some supercomputer time and really be serious. This issue of the weather would be interesting to look into because you are right of course, it directly influences energy demand at given times of the year.

I wonder what they are saying these days? I am too thrifty to pay for it....Can you give us a free hint, tdrive?