The Evolution of Forecast Density Combinations in Economics

Aastveit, Knut Are; Mitchell, James; Ravazzolo, Francesco; Dijk, Herman K. van

doi:10.2139/ssrn.3243003

Cited by 8 publications

(4 citation statements)

References 78 publications

(122 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is an arbitrary choice. A wide variety of weighting schemes can be used to combine candidate forecast models, based on, for instance, the preferences and constraints of the forecaster or the historical performance of the candidate models(Aastveit et al 2018, discuss different choices for density weight estimation). In future work, we intend to explore alternatives as they relate to SAP prediction.…”

mentioning

confidence: 99%

Incorporating Uncertainty into USDA Commodity Price Forecasts

Adjemian

Bruno

Robe

2020

American J Agri Economics

View full text Add to dashboard Cite

From 1977 through April 2019, USDA published monthly season‐average price (SAP) forecasts for key agricultural commodities in the form of intervals meant to indicate forecasters' uncertainty but without attaching a confidence level. In May 2019, USDA eliminated the intervals and began publishing a single point estimate—a value that has a very low probability of being realized. We demonstrate how a density forecasting format can improve the usefulness of USDA price forecasts and explain how such a methodology can be implemented. We simulate 21 years of out‐of‐sample density‐based SAP forecasts using historical data, with forward‐looking, backward‐looking, and composite methods, and we evaluate them based on commonly‐accepted criteria. Each of these approaches would offer USDA the ability to portray richer and more accurate price forecasts than its old intervals or its current single point estimates. Backward‐looking methods require little data and provide significant improvements. For commodities with active derivatives markets, option‐implied volatilities (IVs) can be used to generate forward‐looking and composite models that reflect (and adjust dynamically to) market sentiment about uncertainty—a feature that is not possible using backward‐looking data alone. At certain forecast steps, a composite method that combines forward‐ and backward‐looking information provides useful information regarding farm‐level prices beyond that contained in IVs.

show abstract

mentioning

confidence: 99%

Incorporating Uncertainty into USDA Commodity Price Forecasts

Adjemian

Bruno

Robe

2020

American J Agri Economics

View full text Add to dashboard Cite

show abstract

“…When the true distribution is from the same family as the individual distributions, this shape-preserving property can offer calibration benefits [40]. Other reviews provide further discussion of the theoretical properties of these methods [40,41,51].…”

Section: Aggregation Theory and Methodologiesmentioning

confidence: 99%

Context-dependent representation of within- and between-model uncertainty: aggregating probabilistic predictions in infectious disease epidemiology

Howerton

Runge

Bogich

et al. 2023

J. R. Soc. Interface.

View full text Add to dashboard Cite

Probabilistic predictions support public health planning and decision making, especially in infectious disease emergencies. Aggregating outputs from multiple models yields more robust predictions of outcomes and associated uncertainty. While the selection of an aggregation method can be guided by retrospective performance evaluations, this is not always possible. For example, if predictions are conditional on assumptions about how the future will unfold (e.g. possible interventions), these assumptions may never materialize, precluding any direct comparison between predictions and observations. Here, we summarize literature on aggregating probabilistic predictions, illustrate various methods for infectious disease predictions via simulation, and present a strategy for choosing an aggregation method when empirical validation cannot be used. We focus on the linear opinion pool (LOP) and Vincent average, common methods that make different assumptions about between-prediction uncertainty. We contend that assumptions of the aggregation method should align with a hypothesis about how uncertainty is expressed within and between predictions from different sources. The LOP assumes that between-prediction uncertainty is meaningful and should be retained, while the Vincent average assumes that between-prediction uncertainty is akin to sampling error and should not be preserved. We provide an R package for implementation. Given the rising importance of multi-model infectious disease hubs, our work provides useful guidance on aggregation and a deeper understanding of the benefits and risks of different approaches.

show abstract

“…Except for linear combinations, probabilistic combinations include other techniques such as nonlinear pools (Gneiting & Ranjan, 2013;Ranjan & Gneiting, 2010;Bassetti et al, 2018), Bayesian model averaging (BMA) (Garratt et al, 2003;Moral-Benito, 2015;Aastveit et al, 2018), Bayesian predictive synthesis (BPS) (McAlinn & West, 2019;McAlinn et al, 2020) and quantile forecasting (Lichtendahl Jr et al, 2013;Busetti, 2017;Trapero et al, 2019). These probabilistic combination approaches have specific advantages and shortcomings.…”

Section: Probabilistic Forecasting Combinationsmentioning

confidence: 99%

School Mathematics Textbooks in China

Wang¹

2015

East China Normal University Scientific Reports

View full text Add to dashboard Cite

Let µ and ν be two probability measures on R d , where µ(dx) = e −V (x) dx R d e −V (x) dx for some V ∈ C 1 (R d ). Explicit sufficient conditions on V and ν are presented such that µ * ν satisfies the log-Sobolev, Poincaré and super Poincaré inequalities. In particular, if V (x) = λ|x| 2 for some λ > 0 and ν(e λθ|•| 2 ) < ∞ for some θ > 1, then µ * ν satisfies the log-Sobolev inequality. This improves and extends the recent results on the log-Sobolev inequality derived in [20] for convolutions of the Gaussian measure and compactly supported probability measures. On the other hand, it is well known that the log-Sobolev inequality for µ * ν implies ν(e ε|•| 2 ) < ∞ for some ε > 0. RésuméDes conditions explicites suffisantes sur V et ν sont présentées telles que µ * ν satisfait des inégalités de Sobolev logarithmique, de Poincaré et de super-Poincaré. En particulier, si V (x) = λ|x| 2 pour quelque λ > 0 et ν(e λθ|•| 2 ) < ∞ avec θ > 1, alors µ * ν satisfait l'inégalité de Sobolev logarithmique. Cela améliore et étend des résultats récents sur l'inégalité de Sobolev logarithmique obtenus dans [20] pour des convolutions de la mesure de Gauss et des mesures de probabilité à support compact. D'autre part, il est bien connu que l'inégalité de Sobolev logarithmique pour µ * ν implique ν(e ε|•| 2 ) < ∞ pour quelque ε > 0.

show abstract

The Evolution of Forecast Density Combinations in Economics

Cited by 8 publications

References 78 publications

Incorporating Uncertainty into USDA Commodity Price Forecasts

Incorporating Uncertainty into USDA Commodity Price Forecasts

Context-dependent representation of within- and between-model uncertainty: aggregating probabilistic predictions in infectious disease epidemiology

School Mathematics Textbooks in China

Contact Info

Product

Resources

About