Aggregating Predictions via Sequential Mini-Trading

Mindika Premachandra, Mark Reid
Proceedings of the 5th Asian Conference on Machine Learning, PMLR 29:373-387, 2013.

Abstract

Prediction markets which trade on contracts representing unknown future outcomes are designed specifically to aggregate expert predictions via the market price. While there are some existing machine learning interpretations for the market price and connections to Bayesian updating under the equilibrium analysis of such markets, there is less of an understanding of what the instantaneous price in sequentially traded markets means. In this paper we show that the prices generated in sequentially traded prediction markets are stochastic approximations to the price given by an equilibrium analysis. We do so by showing the equilibrium price is a solution to a stochastic optimisation problem which is solved by stochastic mirror descent (SMD) by a class of sequential pricing mechanisms. This connection leads us to propose a scheme called “mini-trading” which introduces a parameter related to the learning rate in SMD. We prove several properties of this scheme and show that it can improve the stability of prices in sequentially traded prediction markets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v29-Premachandra13, title = {Aggregating Predictions via Sequential Mini-Trading}, author = {Premachandra, Mindika and Reid, Mark}, booktitle = {Proceedings of the 5th Asian Conference on Machine Learning}, pages = {373--387}, year = {2013}, editor = {Ong, Cheng Soon and Ho, Tu Bao}, volume = {29}, series = {Proceedings of Machine Learning Research}, address = {Australian National University, Canberra, Australia}, month = {13--15 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v29/Premachandra13.pdf}, url = {https://proceedings.mlr.press/v29/Premachandra13.html}, abstract = {Prediction markets which trade on contracts representing unknown future outcomes are designed specifically to aggregate expert predictions via the market price. While there are some existing machine learning interpretations for the market price and connections to Bayesian updating under the equilibrium analysis of such markets, there is less of an understanding of what the instantaneous price in sequentially traded markets means. In this paper we show that the prices generated in sequentially traded prediction markets are stochastic approximations to the price given by an equilibrium analysis. We do so by showing the equilibrium price is a solution to a stochastic optimisation problem which is solved by stochastic mirror descent (SMD) by a class of sequential pricing mechanisms. This connection leads us to propose a scheme called “mini-trading” which introduces a parameter related to the learning rate in SMD. We prove several properties of this scheme and show that it can improve the stability of prices in sequentially traded prediction markets.} }
Endnote
%0 Conference Paper %T Aggregating Predictions via Sequential Mini-Trading %A Mindika Premachandra %A Mark Reid %B Proceedings of the 5th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2013 %E Cheng Soon Ong %E Tu Bao Ho %F pmlr-v29-Premachandra13 %I PMLR %P 373--387 %U https://proceedings.mlr.press/v29/Premachandra13.html %V 29 %X Prediction markets which trade on contracts representing unknown future outcomes are designed specifically to aggregate expert predictions via the market price. While there are some existing machine learning interpretations for the market price and connections to Bayesian updating under the equilibrium analysis of such markets, there is less of an understanding of what the instantaneous price in sequentially traded markets means. In this paper we show that the prices generated in sequentially traded prediction markets are stochastic approximations to the price given by an equilibrium analysis. We do so by showing the equilibrium price is a solution to a stochastic optimisation problem which is solved by stochastic mirror descent (SMD) by a class of sequential pricing mechanisms. This connection leads us to propose a scheme called “mini-trading” which introduces a parameter related to the learning rate in SMD. We prove several properties of this scheme and show that it can improve the stability of prices in sequentially traded prediction markets.
RIS
TY - CPAPER TI - Aggregating Predictions via Sequential Mini-Trading AU - Mindika Premachandra AU - Mark Reid BT - Proceedings of the 5th Asian Conference on Machine Learning DA - 2013/10/21 ED - Cheng Soon Ong ED - Tu Bao Ho ID - pmlr-v29-Premachandra13 PB - PMLR DP - Proceedings of Machine Learning Research VL - 29 SP - 373 EP - 387 L1 - http://proceedings.mlr.press/v29/Premachandra13.pdf UR - https://proceedings.mlr.press/v29/Premachandra13.html AB - Prediction markets which trade on contracts representing unknown future outcomes are designed specifically to aggregate expert predictions via the market price. While there are some existing machine learning interpretations for the market price and connections to Bayesian updating under the equilibrium analysis of such markets, there is less of an understanding of what the instantaneous price in sequentially traded markets means. In this paper we show that the prices generated in sequentially traded prediction markets are stochastic approximations to the price given by an equilibrium analysis. We do so by showing the equilibrium price is a solution to a stochastic optimisation problem which is solved by stochastic mirror descent (SMD) by a class of sequential pricing mechanisms. This connection leads us to propose a scheme called “mini-trading” which introduces a parameter related to the learning rate in SMD. We prove several properties of this scheme and show that it can improve the stability of prices in sequentially traded prediction markets. ER -
APA
Premachandra, M. & Reid, M.. (2013). Aggregating Predictions via Sequential Mini-Trading. Proceedings of the 5th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 29:373-387 Available from https://proceedings.mlr.press/v29/Premachandra13.html.

Related Material