INDEPENDENT NEWS

Marsden Fund grants go to Business School economists

Published: Thu 3 Nov 2016 04:14 PM
Marsden Fund grants go to Business School economists
Research at the University of Auckland Business School that could help make online voting more reliable, and a project to improve the accuracy of economic forecasting in the property market have received a total of $1.41 million in this year’s Marsden Fund round.
Drs Steffen Lippert and Simona Fabrizi, two senior lecturers in the Department of Economics, will lead a $705,000 Marsden project to investigate the way we make decisions as a group – for example in a jury, or in political elections – when the reliability of information is ambiguous.
The project’s title is “Beyond the Jury Paradox: Collective Decision-Making without Common Priors”.
“We will develop robust theoretical models of collective decision-making with ambiguous information, and test these models in large-scale laboratory experiments,” says Dr Lippert, who is also the Acting Head of the Department of Economics.
“We know that groups can outperform individuals in many instances of decision-making because a group’s combined information is more reliable than that of any single individual. But when unanimous agreement rather than just a majority of votes is required, group decisions become prone to error because even well-intentioned group members have incentives to disregard information and vote strategically.”
The researchers’ early findings suggest this is because of flawed assumptions about the reliability of information.
“In real life, often it’s not clear whether much of the information we have to go on is reliable, whether as voters or jurors,” says Dr Lippert. “In more robust models of collective decision-making that properly take into account ambiguous information, unanimity may outperform majority voting – resulting in less wrongful convictions, for example.”
Final results will reveal when it’s best to use consensus versus majority voting, and aid in the design of rules for referenda, such as the one used to vote over Brexit, or for digital voting platforms, he says.
Dr Ryan Greenaway-McGrevy, a senior lecturer in the Economics Department, will lead a project that aims to inform better housing policy. His co-collaborator is Distinguished Professor Peter PCB Phillips at the Business School and Sterling Professor of Economics and Professor of Statistics at Yale University.
The title of their project is “New Methods of Panel Data Forecasting Applied to New Zealand’s Property Market”.
Panel data is data collected over time and different units (such as firms, households or regions) – the kind of complex information needed to make sense of the property market. Currently, there are few tools to guide policymakers and others in analysing this vast array of property data.
“We’ll develop techniques that can be used to forecast and analyse property prices, potential contagion across regions, and the effects of proposed policy responses to the housing crisis in Auckland and its spill-over effects on other regions of New Zealand,” says Dr Greenaway-McGrevy.
The researchers will then use those techniques to analyse suggested policy fixes to Auckland’s housing crisis already on the table.
“This Marsden Award is an exciting opportunity for Peter and I to pursue this research, which will help address a pressing problem,” Dr Greenaway-McGrevy says. “We are thrilled to be able to give back to the university, the Business School, and to New Zealand, which have all contributed so much to our lives.
“Our warmest appreciation goes to Professor Greg Whittred and many others at the school for their continued strong support.”
The annual Marsden Fund awards are administered by the Royal Society of New Zealand and allocated over three years.
ends

Next in Business, Science, and Tech

General Practices Begin Issuing Clause 14 Notices In Relation To The NZNO Primary Practice Pay Equity Claim
By: Genpro
Global Screen Industry Unites For Streaming Platform Regulation And Intellectual Property Protections
By: SPADA
View as: DESKTOP | MOBILE © Scoop Media