Extending Extended Logistic Regression: Extended versus Separate versus Ordered versus Censored

Messner, Jakob W.; Mayr, Georg J.; Wilks, Daniel S.; Zeileis, Achim

doi:10.1175/mwr-d-13-00355.1

Cited by 66 publications

(63 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In their native formulations, ELR and HELR result in continuous predictive distributions that are not well suited for variables like local sea ice concentration (SIC), which can have a discrete component at 0 and/or 1. While methods have been developed to extend ELR/HELR to other mixed discrete‐continuous variables (Messner et al, ), such approaches have not yet been extended to SIC forecasts. To account for the discrete component of the SIC distribution, Dirkson et al () recommended using the zero‐ and one‐inflated beta (BEINF) distribution (Ospina & Ferrari, ) for representing the predictive SIC distribution parametrically.…”

Section: Introductionmentioning

confidence: 99%

A Multimodel Approach for Improving Seasonal Probabilistic Forecasts of Regional Arctic Sea Ice

Dirkson

Denis

Merryfield

2019

Geophysical Research Letters

View full text Add to dashboard Cite

We formulate seasonal probabilistic forecasts of Arctic sea ice concentration from a multimodel (MM) ensemble constructed from six state‐of‐the‐art climate models. Trend‐adjusted quantile mapping is applied to postprocess individual model forecasts prior to MM combination, and a comparison is made against two benchmark MM ensembles: one uncorrected and another where individual models are adjusted for mean and trend bias. Focusing on September hindcasts over 2000–2015 initialized monthly from April–August, calibration improves forecast skill for all models tested, but greatest skill is achieved by the calibrated MM ensemble. Compared against a climatology adjusted for trends, skill is seen over most of the Arctic for all MM formulations and at all lead times, with highest skill obtained by the calibrated MM ensemble. Furthermore, despite the overall skill for each MM formulation, we find evidence of the “spring predictability skill barrier” for forecasts initialized before June.

show abstract

Section: Introductionmentioning

confidence: 99%

A Multimodel Approach for Improving Seasonal Probabilistic Forecasts of Regional Arctic Sea Ice

Dirkson

Denis

Merryfield

2019

Geophysical Research Letters

View full text Add to dashboard Cite

show abstract

“…We treat the zero values of simulated and observed flow as censored data, with unknown exact values equal to or below zero. Data censoring approaches that deal with threshold data have been successfully used in a number of forecasting applications (Li et al, 2019;Messner et al, 2014;Scheuerer & Hamill, 2015;Wang & Robertson, 2011).…”

Section: Treatment Of Threshold Datamentioning

confidence: 99%

A Data Censoring Approach for Predictive Error Modeling of Flow in Ephemeral Rivers

Wang

Bennett

Li³

2020

Water Resources Research

View full text Add to dashboard Cite

Flow simulations of ephemeral rivers are often highly uncertain. Therefore, error models that can reliably quantify predictive uncertainty are particularly important. Existing error models are incapable of producing predictive distributions that contain >50% zeros, making them unsuitable for use in highly ephemeral rivers. We propose a new method to produce reliable predictions in highly ephemeral rivers. The method uses data censoring of observed and simulated flow to estimate model parameters by maximum likelihood. Predictive uncertainty is conditioned on the simulation in such a way that it can generate >50% zeros. Our method allows the setting of a censoring threshold above zero. Many conceptual hydrological models can only approach, but never equal, zero. For these hydrological models, we show that setting a censoring threshold slightly above zero is required to produce reliable predictive distributions in highly ephemeral catchments. Our new method allows reliable predictions to be generated even in highly ephemeral catchments.Plain Language Summary Many rivers cease to flow at various times. These rivers are difficult to model well, meaning that the models have a high level of uncertainty. There are no existing methods to correctly quantify the uncertainty in models of rivers that cease to flow >50% of the time. We propose a new method that can quantify the uncertainty of these models, even for rivers that cease to flow very often.Key Points:• Current hydrological error models cannot handle cases where >50% flow is zero • We introduce a new error modeling method that uses data censoring of simulations and observations • Our model produces reliable predictions even in highly ephemeral catchmentsSupporting Information:• Supporting Information S1

show abstract

“…Statistical preprocessing techniques are used to correct such biases (Shukla and Lettenmaier 2011, Yuan and Wood 2012, Crochemore et al 2016, Lucatero et al 2018. We use a logistic regression model to statistically preprocess the CFSv2 forecasts (Messner et al 2014a, 2014b, Yang et al 2017, Sharma et al 2018. This model has been successfully used before with weather forecasts (Sharma et al 2019), and it is tested here for the first time with climate forecasts.…”

Section: Dynamical-statistical Approach To S2s Water Quantity and Quamentioning

confidence: 99%

Seasonal hydroclimatic ensemble forecasts anticipate nutrient and suspended sediment loads using a dynamical-statistical approach

Sharma

Gall

Gironás

et al. 2019

Environ. Res. Lett.

View full text Add to dashboard Cite

Subseasonal-to-seasonal (S2S) water quantity and quality forecasts are needed to support decision and policy making in multiple sectors, e.g. hydropower, agriculture, water supply, and flood control. Traditionally, S2S climate forecasts for hydroclimatic variables (e.g. precipitation) have been characterized by low predictability. Since recent next-generation S2S climate forecasts are generated using improved capabilities (e.g. model physics, assimilation techniques, and spatial resolution), they have the potential to enhance hydroclimatic predictions. Here, this is tested by building and implementing a new dynamical-statistical hydroclimatic ensemble prediction system. Dynamical modeling is used to generate S2S flow predictions, which are then combined with quantile regression to generate water quality forecasts. The system is forced with the latest S2S climate forecasts from the National Oceanic and Atmospheric Administration's Climate Forecast System version 2 to generate biweekly flow, and monthly total nitrogen, total phosphorus, and total suspended sediment loads. By implementing the system along a major tributary of the Chesapeake Bay, the largest estuary in the US, we demonstrate that the dynamical-statistical approach generates skillful flow, nutrient load, and suspended sediment load forecasts at lead times of 1-3 months. Through the dynamical-statistical approach, the system comprises a cost and time effective solution to operational S2S water quality prediction.

show abstract

Extending Extended Logistic Regression: Extended versus Separate versus Ordered versus Censored

Cited by 66 publications

References 27 publications

A Multimodel Approach for Improving Seasonal Probabilistic Forecasts of Regional Arctic Sea Ice

A Multimodel Approach for Improving Seasonal Probabilistic Forecasts of Regional Arctic Sea Ice

A Data Censoring Approach for Predictive Error Modeling of Flow in Ephemeral Rivers

Seasonal hydroclimatic ensemble forecasts anticipate nutrient and suspended sediment loads using a dynamical-statistical approach

Contact Info

Product

Resources

About