Hello gifted R people,
I would like to ask about professional gambling packages in R. I work for a
European gambling syndicate and am interested in the statistical modelling
for sports results to aid investment decision making. The approach is
broadly:
1) Find a predictive model that estimates true probabilities (and odds) more
accurately than the market implied probabilities.
2) Determine risk-neutral prices for various exotic gambling products
(usually involving setting fractional lines).
3) Determine wealth allocation using utility theory to compare investment
products on their price vs probabilistic values. (This is essentially
related the work of Edward Thorp.)
There are several benefits to this asset class:
1) There is no tax payable on gains from gambling in many jurisdictions.
2) Winners can be disguised amongst a large pool of losers through agents to
bookmakers.
3) The market is driven mostly by irrational players.
4) Most recreational players are based in Asia; given the economic growth
and increase in disposable income in the area, the market is expanding
rapidly.
5) Asian bookmakers are inexperienced in pricing risks. As such there are
many counter-parties to take on mis-priced risk.
6) Bets on different events are uncorrelated.
I know of a handful of packages that address this area: bivpois, LearnBayes,
and MCMCpack, but would appreciate insight from the mailing list.
Also, is there interest in forming a R-SIG on modelling professional sports?
Thanks,
Brighton
[[alternative HTML version deleted]]
_______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-finance
-- Subscriber-posting only. If you want to post, subscribe first.
-- Also note that this is not the r-help list where general R questions should
go.
