Skip to main content

Current and Historical Market-Based Probabilities

What are the weights that financial markets assign to a possible event? View our customized data to weigh the options

In making a current decision, how should a policymaker weigh the net benefits of that decision in one possible future against the net benefits of that same decision in another possible future? That’s the essential question asked—and then answered by the data below—by calculating market-based probabilities. Simply put, market-based probabilities are the weights that financial markets assign to a possible event.

Here, we’ve provided a tool that generates customized data on market-based probabilities and then explains the motivation for reviewing such data. We also describe in more detail below the data and calculations used to produce the probabilities, including whether actual transaction prices are used. Note that we use dealer quotes (or the equivalent) when there is not sufficient trading, and potentially even no trading, to provide market prices.



 

Hover or tap on lines to see data point detail

 
 

MPD statistics [.csv]
Statistics for Market-Based Probability Densities (MPDs) generated by the Federal Reserve Bank of Minneapolis.
Data definitions for statistics [.csv]


Market-Based Probabilities: A Tool for Policymakers [PDF]
Ron Feldman, Ken Heinecke, Narayana Kocherlakota, Sam Schulhofer-Wohl and Tom Tallarini


We can observe the current prices of assets (commodities like gold or wheat or financial assets like stocks) from transactions that are taking place every instant in markets all over the world. But the public and policymakers often need to make decisions that depend on what’s going to happen to asset prices in the future. For a large number of assets, we use prices from option markets to estimate the chance or probability of future changes in that asset’s price. Thus, on this web page, we provide estimates of the probability of a 20 percent increase in the S&P 500 over the coming year or the probability of a 20 percent fall in the dollar value of the euro over the next six months.

To be more precise, we provide option market-based estimates of probabilities. We do so because the Federal Reserve Bank of Minneapolis has concluded that the economic policymakers will typically find market-based probabilities useful in their decision-making. Most importantly, the market-based probability accounts for how valuable resources will be in the future relative to today. For example, suppose we find that the market-based probability of a 20 percent fall in real estate prices is larger than the market-based probability of a 20 percent increase in real estate prices. We can conclude that market participants’ current valuation of resources in the former “large decline” case is higher than their current valuation of resources in the latter “large increase” case. Policymakers can best compare current economic costs against future economic benefits (or vice versa) if they make use of this kind of information about the current valuation of future resources. Market-based probabilities are often described by economists as risk-neutral probabilities.


Additional details

Commodities (agriculture, crude oil, and metals)

  • The data used to construct the probabilities are options on futures contracts on the individual crops.
  • End-of-day prices for options that had positive trading volume over the preceding five days are utilized.
  • The technique used to generate the probability distribution is a variation of the procedure described in Shimko (1993).

Equities (banks and stock market indexes)

  • The data used to construct the probabilities are options on the individual stocks or indexes.
  • End-of-day prices for options that had positive trading volume over the preceding five days are utilized.
  • The technique used to generate the probability distribution is a variation of the procedures described in Shimko (1993).

Exchange rates

  • The data used to construct the probabilities are options on futures contracts on the individual exchange rates.
  • End-of-day prices for options that had positive trading volume in the last five days are utilized.
  • The technique used to generate the probability distribution is a variation of the procedures described in Shimko (1993).

Inflation

  • The data used to construct the probabilities are inflation caps and floors.
  • Quotes for the caps and floor contracts are obtained from Bloomberg.
  • The technique used to generate the distributions is a variation of the procedure described in Kitsul and Wright (2012).

Interest rates

  • The data used to construct the probabilities are interest rate caps and floors and options on Treasury futures.
  • Quotes for the caps and floor contracts are obtained from Bloomberg. End-of-day prices for options that had positive trading volume over the preceding five days are used for the Treasury securities.
  • The technique used to generate the distributions is a variation of the procedure described in Kitsul and Wright (2012).

We also utilize several concepts and techniques from other well-established procedures as part of our generation process. Please consult the references listed below for more information about this topic. A more complete explanation of the generation process is provided in the associated methodology document.


References

Breeden, Douglas T., and Litzenberger, Robert H. 1978. Prices of State-Contingent Claims Implicit in Option Prices. Journal of Business 51:4, 621-51.

Figlewski, Stephen. 2008. Estimating the Implied Risk Neutral Density for the U.S. Market Portfolio. In Volatility and Time Series Econometrics: Essays in Honor of Robert F. Engle. Tim Bollerslev, Jeffrey R. Russell, and Mark Watson, eds. Oxford University Press.

Jackwerth, Jens Carsten. 2004. Option-Implied Risk-Neutral Distributions and Risk Aversion, Research Foundation of AIMR.

Kitsul, Yuriy, and Wright, Jonathan H. 2012. The Economics of Options-Implied Inflation Probability Density Functions. Working Paper 18195, National Bureau of Economic Research.

Malz, Allan M. 2014. A Simple and Reliable Way to Compute Option-Based Risk-Neutral Distributions. Staff Report 677, Federal Reserve Bank of New York.

Shimko, David. 1993. Bounds of Probability, Risk 6.