Likelihood-based scoring rules for comparing density forecasts in tails

By Cees G. H. Diks, Valentyn Panchenko, Dick Van Dijk. 2011

161 citations recorded by the source corpus.

Subfields: Demography, Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/32375c9f9c0579a67a5077f5ae4f51210eea60d9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Demography

In Economics and Econometrics

In Finance

Citations in the corpus

Forecast verification for extreme value distributions with an application to probabilistic peak wind prediction (2012, background): “Diks et al. (2011) consider weighted versions of the ignorance score, whereas Gneiting and Ranjan (2011) discuss approaches to weight the CRPS.”
Combining density forecasts using focused scoring rules (2017, unknown): “As discussed in Diks et al. (2011), this second term is necessary, however, to obtain a proper scoring rule.”
Forecast combinations: An over 50-year review (2022, background): “…et al. (2017) extended the idea of optimal combinations but estimated optimal weights by either maximizing the censored likelihood scoring rule (Diks et al., 2011) or minimizing a weighted version of the CRPS, allowing forecasters to limit themselves to a specific region of the target…”
Comparison of non-homogeneous regression models for probabilistic wind speed forecasting (2013, background): “The twCRPS in (6) with w(z) = 1{z ≥ r} thus allows us to simultaneously assess the exceedance probabilities for all thresholds greater or equal to r. Similarly, Diks et al. (2011) propose proper weighted versions of the logarithmic score in (4).”
Generalised Density Forecast Combinations (2015, unknown): “In a related context, albeit for evaluation, Diks et al. (2011) discuss the weighted logarithmic scoring rule, w t ( y t +1 ) log q it ( y t +1 ), where the weight function w t ( y t +1 ) emphasises regions of the density of interest; one possibility, as in (29), is that w t ( y t +1 ) = I ( r s −…”