(no subject)
X-posted from Quants
A market I wasn't aware of before is the curious market for weather.
The chicago stock exchange trade financial instruments based on future weather patterns.
In short, it's a form of insurance or hedging used by suppliers of electricity, which earn or loose a lot of money depending on the weather. The amsterdam airport also buys instruments such as the "frost index" to hedge against the extra costs incurred by runways freezing over.
These instruments have several interesting aspects from a quant point of view. One is spreading your risk, since the weather in Amsterdam is completely independent of the stock market. There is no intricate web that links house prices in ohio to heatwaves in florida.
Another aspect is a peculiar thing called "spatial finance". If a local energy company in Copenhagen wants to buy a temperature derivate for Copenhagen, it can't. The Chicago stock market only sell it for 10 european cities. The solution is to buy a mix of different cities. For instance Oslo, Stockholm, Berlin and Amsteram, which roughly circle Copenhagen. How to calculate the weighting is a process that mixes the methods of the quant with meteorology.
A third aspect, which professor Fred Espen Benth at the university of Oslo, Norway, has focused on is that weather derivates have a memory. Ordinary stocks are said to have a marcovian property in that the stock price will drop like a rock if it gets known that a company is going bust. However, the value of a temperature derivate for a given day becomes less and less uncertain as that day draws near. Weather predictions play a part in this, but there's also a price stability in that temperature will statistically change less over one day than over one week and thus changes less and less running up to the stipulated date. The "temperature stock" is thus restricted from dropping like a rock or to instantly skyrocket.
Especially the memory effect makes this a fascinating financial instrument. In the midst of this, the professor's team has faced a mystery: Since the consumption of electricity and natural gas is strongly dependent on the weather, we should have seen a similar memory effect in the prices of these commodities. But the group at the Centre of Mathematics for Applications has failed to find such an effect in these markets. Is the effect on el and gas prices swamped by other effects, or are traders in these commodities simply not making good enough use of the temperature dependency?
A market I wasn't aware of before is the curious market for weather.
The chicago stock exchange trade financial instruments based on future weather patterns.
In short, it's a form of insurance or hedging used by suppliers of electricity, which earn or loose a lot of money depending on the weather. The amsterdam airport also buys instruments such as the "frost index" to hedge against the extra costs incurred by runways freezing over.
These instruments have several interesting aspects from a quant point of view. One is spreading your risk, since the weather in Amsterdam is completely independent of the stock market. There is no intricate web that links house prices in ohio to heatwaves in florida.
Another aspect is a peculiar thing called "spatial finance". If a local energy company in Copenhagen wants to buy a temperature derivate for Copenhagen, it can't. The Chicago stock market only sell it for 10 european cities. The solution is to buy a mix of different cities. For instance Oslo, Stockholm, Berlin and Amsteram, which roughly circle Copenhagen. How to calculate the weighting is a process that mixes the methods of the quant with meteorology.
A third aspect, which professor Fred Espen Benth at the university of Oslo, Norway, has focused on is that weather derivates have a memory. Ordinary stocks are said to have a marcovian property in that the stock price will drop like a rock if it gets known that a company is going bust. However, the value of a temperature derivate for a given day becomes less and less uncertain as that day draws near. Weather predictions play a part in this, but there's also a price stability in that temperature will statistically change less over one day than over one week and thus changes less and less running up to the stipulated date. The "temperature stock" is thus restricted from dropping like a rock or to instantly skyrocket.
Especially the memory effect makes this a fascinating financial instrument. In the midst of this, the professor's team has faced a mystery: Since the consumption of electricity and natural gas is strongly dependent on the weather, we should have seen a similar memory effect in the prices of these commodities. But the group at the Centre of Mathematics for Applications has failed to find such an effect in these markets. Is the effect on el and gas prices swamped by other effects, or are traders in these commodities simply not making good enough use of the temperature dependency?