Bubbles are to free-market capitalism as hurricanes are to weather. —  Henry Blodget.

Behrend Memorial Lecture at University of Melbourne

Following neatly and accidentally on the theme of uncertainty in decision-making, on Tuesday October 31, Professor Mark Burgman gave a wonderful presentation on Making use of uncertainty in environmental decision-making for the 20th Behrend Memorial Lecture. Mark focussed on the assessment of uncertainty and its role in risk-taking. We were treated to a first-rate demonstration of the principle because the fire-alarm went off just at the conclusion of Mark’s presentation, and the hosts of the evening had to solve a problem in applied decision theory with a fairly tight deadline!
Mark’s abstract was:

Uncertainty is pervasive in environmental decision making. Taxonomies of uncertainty are useful to describe the things we don’t know. Typically, in the past, they have been too narrow to deal with the full range of uncertainty that confronts environmental decision-makers. Lack of knowledge about how systems work and language-based misunderstandings are difficult to describe formally and usually go unacknowledged. This presentation provides some examples and shows their effects in routine environmental issues. It emphasises that ignoring these sources of uncertainty doesn’t make them go away. It outlines some solutions, including satisfactory (rather than optimal) decisions and a more complete characterisation of unknowns. These approaches provide new solutions to previously intractable social dilemmas regarding decisions among competing demands for scarce natural resources.

Here’s a charming photo of Mark:

Mark Burgman

You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

AddThis Social Bookmark Button

One Response to “Behrend Memorial Lecture at University of Melbourne”

  1. I do not know much about how this kind of non-probabilistic uncertainty is handled.

    One method I have seen is worst case. This would seem to be incredibly conservative as there is no central limiting - the possibility of every uncertain variable being at the limit of its range is incredibly remote.

Leave a Reply