Biology Reference
In-Depth Information
the market still demonstrates that the traders collectively thought that novel
influenza A (H1N1) would travel quickly around the world and would affect
a very large number of countries. Nevertheless, this example shows how
volatile prices in lightly traded markets can be and signals the importance of
carefully considering dramatic price changes caused by a very limited num-
ber of trades. In some situations, in thinly traded markets, it might be more
accurate to look at moving-average prices. For example, a “three-day moving
average” or a weekly average (if most traders only trade once per week).
Ultimately, though, our pilot novel influenza A (H1N1) market did show
that a diverse group of participants could accurately predict a series of novel
influenza A (H1N1)-related events using a prediction market based on a
valueless currency. Thus, we think that prediction markets may provide a
flexible and effective way to aggregate both objective and subjective infor-
mation about emerging infectious diseases, even without monetary incen-
tives. Given a pool of informed participants willing to trade in a market, one
can envision that probabilities generated by the market could help public
health officials plan for the future and coordinate resources.
8.5 Conclusions
Prediction markets might seem like a very academic or impractical
approach to forecasting disease activity. However, consider the alternative.
Currently, the public health community does not have any convenient way
to ask a diverse group of experts about the likelihood of the occurrence of
disease-related events. Public health officials can ask their co-workers ques-
tions or sponsor conference calls, but these are difficult to arrange across
different time zones. Surveys could also be distributed, but there are several
limitations to running surveys to aggregate information about emerging
infectious diseases. In fact, prediction markets have several advantages over
surveys: (1) They are continuous and ongoing, allowing immediate revela-
tion of new information. (2) While some surveys offer a small incentive in
return for participation, the incentives earned by traders in a prediction
market increase in proportion to the quality of the information provided.
(3) Unlike surveys, a market provides immediate feedback to participants,
allowing them opportunities to reassess their own information and to
respond. (4) The market interface is interactive, in marked contrast to most
surveys, providing further incentives for participation. (5) Furthermore,
most surveys rely on random samples for validity and accuracy. In predic-
tion markets, on the other hand, the best participants are those with the
best information, and those are the very individuals who are most likely
to self-select into the market. With surveys, this process would introduce a
