FISH 506B (3 credits; Section 001)
Course Organizers: Dr
Murdoch McAllister, Fisheries Centre, UBC
Schedule: Term 2, Tuesday and Thursday 10am-12 noon
Place: Room 320, Aquatic Ecosystems Research Laboratory
(AERL), 2202 Main Mall
Prerequisites and Restrictions:
Minimal entry requirement: first year undergraduate calculus and reasonably good quantitative skills such as those gained through FISH 504 and FISH 505. Students wishing to take the course should also have good computing skills and be able to use Excel spreadsheets to implement simple time series models, apply Monte Carlo simulation methods and estimate the parameters of non-linear models. The course is to be limited to 15 participants.
Course Description:
This course explores the use of Bayesian decision analysis as a quantitative technique with which to inform decision makers about the extent to which alternative decision options may enable them to achieve their objectives, taking into account available information and uncertainty over factors that affect the outcomes of interest. Students are to learn about the conceptual framework for Bayesian decision analysis, attitudes to risk and uncertainty, risk averse and other types of utility functions, minimax and maximin regret and other types of decision making criteria, the concept of expected value of perfect information, different approaches to assigning probabilities to alternative hypotheses, including Bayesian statistical methods, different software options for Bayesian decision analysis calculations, approaches to communicating results obtained from Bayesian decision analysis, the roles of decision analysis results in the making of decisions, and the advantages and limitations of Bayesian decision analysis as an approach to facilitate the use of science in resource management and policy decision making. Lectures and demonstrations will be supplemented with practical sessions, using Excel, WinBUGS, Visual Basic software, and HUGIN software, where students are to work on set problems. Fisheries estimation and decision analysis modeling problems will be explored to highlight the generic features of Bayesian decision analysis and place the approach in the context of on-the-ground management decision making situations..
Preparation:
Weekly recommended readings. Completion of weekly practicals outside of the
scheduled practical each week. There is no required textbook.
Assessment:
The course grade will be made up of three components: (1) Four written assignments, each counting for 20% of the final grade: The first three assignments will be deal with conceptual and methodological aspects of decision analysis. The fourth assignment will involve the application of Bayesian decision analysis to a problem of the student’s choice; (2) one seminar presentation summarizing the student’s application of a Bayesian decision analysis for 10% of the final grade; and (3) a 10 minute seminar presentation in which the student summarizes a peer-reviewed published article dealing with the application of Bayesian decision analysis, counting for 10% of the final grade
Submission deadlines for written work:The deadline dates for written work are to be two weeks following distribution of the problem set. Note that any written work submitted after the deadlines will be subjected to a 5% penalty per day.
Tentative order of Lectures and discussion sessions:
| 1 |
Overview of the course and various generic approaches to dealing with risk and uncertainty in decision making |
| 2 |
Exploring the conceptual framework of Bayesian decision analysis |
| 3 |
Exploring attitudes to uncertainty, risk averse and other types of utility functions and approaches to formulating utility functions |
| 4 |
Exploring minimax and maximin regret criteria for choosing among alternative actions that can be taken, the effect of discounting on the perception of the merits of alternative actions, and the expected value of perfect information |
| 5 |
Exploring the concept of expected value of perfect information |
| 6 |
Exploring alternative approaches to assigning probabilities to alternative hypotheses, with focus on Bayesian methods for parameter estimation |
| 7 |
Identifying and formulating alternative management options to be evaluated |
| 8 |
Reviewing implementations of Bayesian decision analysis |
| 9 |
Reviewing the software options available for Bayesian decision analysis |
| 10 |
Reviewing successes and failures in the communication of decision analysis results |
| 11 |
Evaluating the benefits and limitations of Bayesian decision analysis methods |
| 12 |
Reviewing the main themes of the course and concluding remarks |
Topics for the practical sessions:
| 1 |
Implementing a simple quantitative decision analysis in
a spreadsheet |
| 2 |
Modeling the potential consequences of alternative actions using alternative parameter values in a single model and outputting results in decision tables |
| 3 |
Exploring the impacts of different types of utility functions on decision analysis results |
| 4 |
Exploring minimax and maximin regret criteria for choosing among alternative actions that can be taken, the effect of discounting on the perception of the merits of alternative actions, and the expected value of perfect information |
| 5 |
Using WinBUGS to estimate parameters of systems dynamics models |
| 6 |
Using WinBUGS to compute marginal posteriors for alternative systems dynamics models |
| 7 |
Using WinBUGS to do stochastic projections |
| 8 |
Learning to program a simple fisheries dynamics model in Visual Basic |
| 9 |
Doing stochastic open loop fisheries model projections in Visual Basic |
| 10 |
Applying stochastic feedback control policies in Visual Basic |
| 11 |
Learning to formulate decision analysis problems using HUGIN decision analysis software |
| 12 |
Learning about Bayesian inference and the value of information with the use of Bayesian networks and HUGIN software |
| 13 |
Considering the impact of the formulation of decision analysis models and alternative utility functions on the perception of optimal decisions using HUGIN software |
Recommended readings:
There is as yet no single textbook chosen for this course. Some recommended
journal article and book chapter readings include the following:
Ellison, A.M. 1996. An introduction to Bayesian inference for ecological
research and environmental decision-making. Ecological applications, 6, 1036-1046.
Francis, R. I. C. C., and R. Shotton. 1997. "Risk" in fisheries
management: A review. Canadian Journal of Fisheries and Aquatic Sciences 54(8):1699-1715.
Hilborn, R. and C. J. Walters. 1992. Quantitative Fisheries Stock Assessment:
Choice, Dynamics and Uncertainty. Chapman and Hall, New York. 570 p. Chapters
1-4, 15-18.
Hilborn, R., Pikitch, E.K., and Francis, R.C. 1993. Current trends in including
risk and uncertainty in stock assessment and harvest decisions. Canadian Journal
of Fisheries and Aquatic Sciences 50, 874-880.
Hill, S., L., Watters, G.M., Punt, A.E., McAllister, M.K., LeQuere, C., Turner,
J. Model uncertainty in the ecosystem approach to fisheries. Submitted to
Fish and Fisheries.
Jacobson, L.D., and Cadrin, S.X. 2003. Stock-rebuilding time isopleths and
constant-F stock rebuilding plans for overfished stocks. Fishery Bulletin.
100: 519-539.
Keeny, R. and Raiffa, H. 1976. Decisions with Multiple Objectives: Preferences
and Value Tradeoffs. J. Wiley, New York.
Lane, D. E., and R. L. Stephenson. 1998. A framework for risk analysis in
fisheries decision-making. ICES J.Mar.Sci 55(1):1-13.
McAllister, M.K. and Kirkwood, G.P. "Bayesian stock assessment: a review
and example application using the logistic model" ICES J. Mar. Sci. 55,
(1998), 1031-1060.
McAllister, M.K. and Kirkwood, G.P. "Using Bayesian decision analysis
to help achieve a precautionary approach to managing newly developing fisheries"
Can. J. Fish. Aquat. Sci. 55, (1998), 2642-2661.
McAllister, M. K., E. K. Pikitch, A. E. Punt, and R. Hilborn. 1994. A Bayesian
approach to stock assessment and harvest decisions using the Sampling/ Importance
Resampling Algorithm. Canadian Journal of Fisheries and Aquatic Science 51:
2673-2687.
McAllister, M.K. Pikitch, E.K., and Babcock, E.A. "Using demographic
methods to construct Bayesian priors for the intrinsic rate of increase in
the Schaefer model and implications for stock rebuilding." Can. J. Fish.
Aquat. Sci. 58(9) (2001): 1871-1890.
McAllister, M.K., P. J. Starr, V. Restrepo and G. P. Kirkwood. 1999. Formulating
quantitative methods to evaluate fishery management systems: what fishery
processes should we model and what trade-offs do we make? ICES Journal of
Marine Science. 56: 900-916.
Patterson, K., R. Cook, C. Darby, S. Gavaris, L. Kell, P. Lewy, B. Mesnil,
A. Punt, V. Restrepo, D. W. Skagen, and G. Stefánsson. 2001. Estimating
uncertainty in fish stock assessment and forecasting. Fish and Fisheries 2
(2), 125-157.
Punt, A.E. and R. Hilborn. 1997. Fisheries stock assessment and decision analysis:
A review of the Bayesian approach. Rev. Fish. Biol. Fish. 7: 35-63.
Raiffa, H. 1968. Decision Analysis. London: Addison-Wesley.
Restrepo, V.R., Hoenig, J.M., Powers, J.E., Baird, J.W., and Turner, S.C.
1992. A simple simulation approach to risk and cost analysis, with application
to swordfish and cod fisheries. Fishery Bulletin 90: 736-48.
Robb, C. A., and Peterman, R. M. 1998. Application of Bayesian decision analysis
to the management of a sockeye salmon fishery. Canadian Journal of Fisheries
and Aquatic Sciences, 55: 86-98.
Sainsbury, K. 1988. The ecological basis of multispecies fisheries, and management
of a demsersal fishery in tropical Australia. In Population dynamics (2nd
edn.). pp. 349-382. Ed. by J. A. Gulland. John Wiley and Sons Ltd. pp. 422
Wade, P.R. (2000). Bayesian methods in conservation biology. Conservation
Biology, 14, 1308-1316.
Walters, C., and J.-J. Maguire. 1996. Lessons for stock assessment from the
northern cod collapse. Reviews in Fish Biology and Fisheries 6: 125-137.
Walters, C.,J., and Martell, S.J. 2004. Fisheries ecology and management.
Princeton University Press, Princeton. Chapters 1-5.
