[R-SIG-Finance] commodities futures

[David Lüthi]

Hi Edouard

There is an R package which deals with commodity futures. It's name is
schwartz97 and can be found on R-forge under the Rmetrics project (it
will be on CRAN soon). As revealed by its name, this package
implements the 2 factor model of Eduardo Schwartz.

In general commodity modelling is not straight-forward at all and a
commodity model per se does not exist. The characteristics of
different commodities can be completely different.  Gold, e.g.,
behaves as a financial asset and therefore standard term-structure
models can be used. Flow commodities as electricity can exhibit very
wild behavior as zero prices and very steep forward curves in both
directions. The spot market (if it exists) is likely to be illiquid
but futures are traded in high volumes. Hence, the dynamics of the
forward curve is the object of interest and the plain spot price
dynamics is least relevant.  But a flexible forward curve implies the
modelling of several factors. The likelihood to run into serious
overfitting when several factors and their interdependence is modelled
is high...

To sum up, the model or heuristic of choice depends heavily on the
commodity and the goal of the modelling. The 2-factor model of
Schwartz is quite appropriate for base metals and some agriculturals
as soybeans and corn if it is about risk management.

Cheers, david


Edouard Tallent wrote:
> hi everyone.
> 
> is there any special R package to deal with commodities futures ?
> 
> anyone interested in sharing knowledge, experience on the implementation of 
> trading strategies on commodities futures markets (energy, softs) with R ?
> 
> cheers,
> edouard.
> 
> "spreads for life"
> 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> R-SIG-Finance at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-finance
> -- Subscriber-posting only. If you want to post, subscribe first.
> -- Also note that this is not the r-help list where general R questions should go.