Making Money from the Weather

There is a nifty new paper (in PDF format) that describes how you can fruitfully use simple time series U.S. weather data to make in money in weather markets:

We take a simple time-series approach to modeling and forecasting daily average temperature in U.S. cities, and we inquire systematically as to whether it may prove useful from the vantage point of participants in the weather derivatives market. The answer is, perhaps surprisingly, yes. Time-series modeling reveals conditional mean dynamics, and crucially, strong conditional variance dynamics, in daily average temperature, and it reveals sharp differences between the distribution of temperature and the distribution of temperature surprises. As we argue, it also holds promise for producing the long-horizon predictive densities crucial for pricing weather derivatives, so that additional inquiry into time-series weather forecasting methods will likely prove useful in weather derivatives contexts.

This is just another example of how new sources of data being exposed by the Internet and elsewhere are creating new investment opportunities. The proof: People are having such a hard time keeping up that you can exploit simple time-series data in a liquid market.

Related posts:

  1. Peer-to-Peer Weather
  2. Making Money the Moneyball way
  3. The Economics of Weather
  4. Peer-to-Peer Weather, Redux
  5. Nasty IPO Weather Out There

Comments

  1. Weather Derivatives

    Via Infectious Greed:
    There is a nifty new paper
    (in PDF format) that describes how you can fruitfully use simple time
    series U.S. weather data to make in money in weather markets:
    We take a simple time-series approach to modeling and forecasting…

  2. Marc Shivers says:

    Paul,
    Campbell & Diebold is one of the better papers I’ve come accross on time-series estimation of weather derivative prices. That said, I have to disagree with your conclusion regarding investment opportunities. Weather derivative markets are actually not very liquid at all, and bid-ask spreads are fairly wide (a 5% spread on CME is not unusual). Furthermore, weather time-series are non-stationary in ways that are not well understood (witness the back-and-forth for the last 20 years on the global warming debate), so you’re likely to have to live with significant model risk over and above the error bounds in your parameter estimates. It’s a very tough problem.