Estimating Jump Diffusion Parameters for Electricity Prices

Summary

This discussion concerns estimating parameters for an Ornstein–Uhlenbeck electricity price model with jumps. The questioner has implemented linear regression and maximum likelihood methods described in a paper on swing options and electricity derivatives, and asks for broader data to benchmark the estimates across markets and periods.

The reply points to New Zealand as a source of public power market data, naming an exchange for live forward prices and a market authority for historical data. It also cautions that high-frequency observations may be difficult to obtain. The post gives no sample data, benchmark estimates, or comparison of the two estimation methods, so it serves mainly as a pointer to data sources rather than evidence about estimator performance.

Key ideas

• The post concerns linear regression and maximum likelihood estimation for an electricity price Ornstein–Uhlenbeck process with jumps.
• The suggested data sources cover New Zealand market forwards and historical electricity market data.
• The discussion provides no benchmark estimates or empirical comparison of estimation methods.
• High-frequency electricity data may be difficult to obtain.

Tags

# Electric power price parameter estimation


# Electric power price parameter estimation












currently I am working through the paper of Tino Kluge "Pricing Swing Options and other Electricity Derivatives" to get a better understanding about the power markets.

The author establishes methods in order to estimate relevant parameters for a ornstein-uhlenbeck process with jumps in chapter 3.3 by the means of linear regression and maximum likelihood.

I implemented the formulas and results seem sensible, but does anyone of you know more data sources than the small table on page 31 of the paper, where I could get benchmark results for more diverse time periods and markets.

Thanks

## Answer by Alexander McFarlane (score 1)

https://quant.stackexchange.com/a/36007

I found New Zealand has some great open source data.

The ASX publishes live forwards:

- http://www.asx.com.au/asx/markets/futuresPriceList.do?code=ED&type=FUTURE

You can also get historic data from the following authority:

- https://www.emi.ea.govt.nz/

If you are after high freq data I'm guessing you may struggle

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.