Blog Archives

Modeling Residential Electricity Usage with R – Part 2

July 8, 2013
By
Modeling Residential Electricity Usage with R – Part 2

(This article was first published on Commodity Stat Arb, and kindly contributed to R-bloggers) I can’t believe it has been nearly 6 months since I last posted.  Given the sustained heat it seemed like a good idea to finish off this subject.As hinted at in my last post, temperature is the missing variable to make sense of Residential electrical...

Read more »

Modeling Residential Electricity Usage with R

January 30, 2013
By
Modeling Residential Electricity Usage with R

Wow, I can’t believe it has been 11 months since my last blog posting!  The next series of postings will be related to the retail energy field.  Residential power usage is satisfying to model as it can be forecast fairly accurately with the right inputs.  Partly as a consequence of deregulation there is now more data more available than...

Read more »

Futures price prediction using the order book data

March 5, 2012
By
Futures price prediction using the order book data

It has been a couple of months since my last post; busy with lots of projects.I had some fun playing around with data from Interactive Brokers API.  It turns out that it is relatively easily to get hold of the raw market data relating to both trades and order book changes for CME/NYMEX commodity futures.  For the purposes of...

Read more »

NG Spreads returns, a reliable earner.

December 1, 2011
By
NG Spreads returns, a reliable earner.

Retail Commodity Investors – A Bleeding Indicator?

May 10, 2011
By
Retail Commodity Investors – A Bleeding Indicator?

Commodities vs. commodity stocks

May 2, 2011
By
Commodities vs. commodity stocks

Platinum – Palladium relationship

January 11, 2011
By
Platinum – Palladium relationship

Oil – Natural Gas Cointegration – turning point?

December 22, 2010
By
Oil – Natural Gas Cointegration – turning point?

Its 9am, do you know what the traders are thinking?

November 17, 2010
By
Its 9am, do you know what the traders are thinking?

Roll proposed a model for the bid-ask spread that was based on first-order serial correlation.  His empirical tests were based on daily and weekly frequency equity data, and based on the results he concluded there were informational inefficiencies (or that there was very short term non-stationarity in expected returns).More recently this model has been applied to high...

Read more »

The curious case of Oct-Jan NG spreads

October 18, 2010
By
The curious case of Oct-Jan NG spreads