Is it Better to Average Probabilities or Quantiles?

Lichtendahl, Kenneth C.; Grushka‐Cockayne, Yael; Winkler, Robert L.

doi:10.2139/ssrn.2066806

Cited by 33 publications

(58 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Obviously, one advantage is that it is easy to understand and compute, but ease of use is not worth much unless the resulting averages perform well. Using strictly proper scoring rules, Lichtendahl et al . show that the simple average of probability forecasts will always score as well as the average of the scores of the individual probability forecasts being averaged.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Equal Versus Differential Weighting in Combining Forecasts

Winkler

2014

Risk Analysis

Self Cite

View full text Add to dashboard Cite

mentioning

confidence: 99%

“…At a minimum, the degree of overconfidence can vary from situation to situation and from variable to variable, even for the same expert, as well as from expert to expert. Lichtendahl et al . found that the existence and degree of overconfidence differed for expert forecasts of GDP growth and inflation and for different lead times for each variable.…”

mentioning

confidence: 99%

Equal Versus Differential Weighting in Combining Forecasts

Winkler

2014

Risk Analysis

Self Cite

View full text Add to dashboard Cite

“…In this thesis we choose to use linear opinion pooling for its relative simplicity, its ease of interpretation and in sticking with common practice. We comment that Litchendahl et al (2013) discusses linear averaging of quantiles rather than probabilities, and some motivations for why this area merits further exploration. In what follows we assume that our linear pools are comprised of probability distributions.…”

Section: Combining Expert Judgmentsmentioning

confidence: 99%

Reliability updating in linear opinion pooling for multiple decision makers

Bolger

Houlding

2016

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

Accurate information sources are vital prerequisites for good decision making. In this thesis we consider a multiple participant setting, where all decision makers (DMs) have a collection of neighbours with whom they share their beliefs about some common relevant uncertain quantity. When determining which course of action to follow a DM takes into account all the information received from her neighbours. Over time, in light of the returns observed from choices made, DMs update their own beliefs over the uncertain event, and also adjust the degree of consideration that they afford to the opinions of each neighbour based on the level of reliability that the information they provide is ascertained to have. Much of this thesis is concerned with constructing a method that incorporates both of these learning facets in a dynamic fashion. This technique, termed the Plug-in approach, is motivated and derived, and attempts are made to justify its use by consideration of some attractive properties it obeys, in addition to studies conducted using both simulated and real data which compared its performance to some rational alternatives. Generalisations of this method are also provided to a setting where DMs specify their opinions nonparametrically rather than using probability distributions, as well as in a group setting where utilities as well as opinions must be amalgamated. Two subjective approaches are also briefly discussed, before we conclude with numerous suggestions for further research in this field.

show abstract

“…This question has been approached in the literature in a number of different ways. The most traditional approach is to aggregate beliefs to produce a single, portable probability distribution (see (Clemen and Winkler 1999) (Cooke and Goossens 2008) (Hora et al 2013) (Lichtendahl, Grushka-Cockayne, and Winkler 2013) for discussions of aggregation methods). The resulting distribution can then be used in a Subjective Expected Utility (SEU) framework (See for example (Baker and Solak 2014;Kelly and Kolstad 1999;Keller, Bolker, and Bradford 2004)).…”

Section: Introductionmentioning

confidence: 99%

Finding Common Ground When Experts Disagree: Robust Portfolio Decision Analysis

2017

View full text Add to dashboard Cite

Within the IEB framework, the Chair of Energy Sustainability promotes research into the production, supply and use of the energy needed to maintain social welfare and development, placing special emphasis on economic, environmental and social aspects. There are three main research areas of interest within the program: energy sustainability, competition and consumers, and energy firms. The energy sustainability research area covers topics as energy efficiency, CO2 capture and storage, R+D in energy, green certificate markets, smart grids and meters, green energy and biofuels. The competition and consumers area is oriented to research on wholesale markets, retail markets, regulation, competition and consumers. The research area on energy firms is devoted to the analysis of business strategies, social and corporative responsibility, and industrial organization. Disseminating research outputs to a broad audience is an important objective of the program, whose results must be relevant both at national and international level.The Chair of Energy Sustainability of the University of Barcelona-IEB is funded by the following enterprises ACS, CEPSA, CLH, Enagas, Endesa, FCC Energia, HC Energia, Gas Natural Fenosa, and Repsol) through FUNSEAM (Foundation for Energy and Environmental Sustainability). introducing an approach we call Robust Portfolio Decision Analysis. We introduce the idea of Belief Dominance as a prescriptive operationalization of a concept that has appeared in the literature under a number of names. We use this concept to derive a set of non-dominated portfolios; and then identify robust individual alternatives from the non-dominated portfolios. The Belief Dominance concept allows us to synthesize multiple conflicting sources of information by uncovering the range of alternatives that are intelligent responses to the range of beliefs. This goes beyond solutions that are optimal for any specific set of beliefs to uncover defensible solutions that may not otherwise be revealed. We illustrate our approach using a problem in the climate change and energy policy context: choosing among clean energy technology R&D portfolios. We demonstrate how the Belief Dominance concept can uncover portfolios that would otherwise remain hidden and identify robust individual investments.

show abstract

Is it Better to Average Probabilities or Quantiles?

Cited by 33 publications

References 22 publications

Equal Versus Differential Weighting in Combining Forecasts

Equal Versus Differential Weighting in Combining Forecasts

Reliability updating in linear opinion pooling for multiple decision makers

Finding Common Ground When Experts Disagree: Robust Portfolio Decision Analysis

Contact Info

Product

Resources

About