Verification tools for probabilistic forecasts of continuous hydrological variables

By Francesco Laio, S. Tamea. 2006

396 citations recorded by the source corpus.

Abstract. In the present paper we describe some methods for verifying and evaluating probabilistic forecasts of hydrological variables. We propose an extension to continuous-valued variables of a verification method originated in the meteorological literature for the analysis of binary variables, and based on the use of a suitable cost-loss function to evaluate the quality of the forecasts. We find that this procedure is useful and reliable when it is complemented with other verification tools, borrowed from the economic literature, which are addressed to verify the statistical correctness of the probabilistic forecast. We illustrate our findings with a detailed application to the evaluation of probabilistic and deterministic forecasts of hourly discharge values.

Subfields: Economics and Econometrics, Finance, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/c9cdd9f326fbeb6407df0fd61a1430ad09a5b584

How researchers see this paper

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

In Economics and Econometrics

In Finance

In Global and Planetary Change

Citations in the corpus

Of quantiles and expectiles: consistent scoring functions, Choquet representations and forecast rankings (2015, background): “Related ideas have recently been put forward in the hydrologic and meteorological literatures (Laio and Tamea, 2007; Bradley and Schwartz, 2011; Smet et al., 2012).”
A Bayesian joint probability modeling approach for seasonal forecasting of streamflows at multiple sites (2009, methodology): “[24] There are many facets to the verification of probabilistic forecasts of a continuous variable [e.g., Gneiting et al., 2007; Laio and Tamea, 2007 ].”
Critical evaluation of parameter consistency and predictive uncertainty in hydrological modeling: A case study using Bayesian total error analysis (2009, neutral): “Res., 45, W00B14, doi:10.1029/2008WR006825.”
Forecast verification for extreme value distributions with an application to probabilistic peak wind prediction (2012, background): “//d (6) see Laio and Tamea (2007), Gneiting and Raftery (2007), and Gneiting and Ranjan (2011).”
Comparison of joint versus postprocessor approaches for hydrological uncertainty estimation accounting for error autocorrelation and heteroscedasticity (2014, methodology): “In this study, reliability is assessed graphically using predictive quantile-quantile (PQQ) plots [Laio and Tamea, 2007; Thyer et al., 2009] and quantified numerically using the metric pReliab½~QkQ presented in Appendix B.”
Characterising performance of environmental models (2013, background): “While not covered in this paper, the reader is referred to a number of useful reviews (Laio and Tamea, 2007; Gneiting and Raftery, 2007; Murphy and Winkler, 1987; Christoffersen, 1998; Boucher et al., 2009).”
Multisite probabilistic forecasting of seasonal flows for streams with zero value occurrences (2011, methodology): “There are many facets to the verification of probabilistic forecasts of a continuous variable [e.g., Gneiting et al., 2007; Laio and Tamea, 2007].”
Kullback–Leibler Divergence as a Forecast Skill Score with Classic Reliability–Resolution–Uncertainty Decomposition (2010, background): “Therefore, skill scores for ordinal category forecasts are often required to be ‘‘sensitive to distance’’ (Epstein 1969; Laio and Tamea 2007; Murphy 1971, 1970).”
Bayesian structured additive distributional regression with an application to regional income inequality in Germany (2015, neutral): “with Gaussian errors εd ∼ N(0, τ 2) and noninformative priors for β1 or β1 and β2 [Lang and Brezger (2004)].”
Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, background): “CRPS can be defined [108] to summarize all possible quantile losses as”
Evaluating Probabilistic Forecasts with scoringRules (2017, methodology): “In the scoringRules package, we use an algebraically equivalent representation of the CRPS based on the generalized quantile function (Laio and Tamea 2007), leading to CRPS(F̂m, y) = 2 m2 m”