2017
DOI: 10.1002/ece3.3621
|View full text |Cite
|
Sign up to set email alerts
|

A hierarchical spatiotemporal analog forecasting model for count data

Abstract: Analog forecasting is a mechanism‐free nonlinear method that forecasts a system forward in time by examining how past states deemed similar to the current state moved forward. Previous applications of analog forecasting has been successful at producing robust forecasts for a variety of ecological and physical processes, but it has typically been presented in an empirical or heuristic procedure, rather than as a formal statistical model. The methodology presented here extends the model‐based analog method of Mc… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
9
1

Year Published

2018
2018
2020
2020

Publication Types

Select...
4
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(10 citation statements)
references
References 45 publications
0
9
1
Order By: Relevance
“…In this section, we describe how the similarities between climate forcings at different times can be used to model the temporal dependence in the offset term. Following McDermott and Wikle (2016) and McDermott et al (2018), who take a similar approach to forecasting soil moisture and waterfowl settling behavior, we will refer to this as an analogue prior on γ t , t = 1, . .…”
Section: Analogue Priormentioning
confidence: 99%
See 1 more Smart Citation
“…In this section, we describe how the similarities between climate forcings at different times can be used to model the temporal dependence in the offset term. Following McDermott and Wikle (2016) and McDermott et al (2018), who take a similar approach to forecasting soil moisture and waterfowl settling behavior, we will refer to this as an analogue prior on γ t , t = 1, . .…”
Section: Analogue Priormentioning
confidence: 99%
“…Analogue methods were originally developed as empirical tools for short-term weather forecasting (Krick, 1942) and climate modeling (Barnett and Preisendorfer, 1978), but researchers have begun to recognize their utility in a variety of contexts, including machine learning (Zhao and Giannakis, 2016;Lguensat et al, 2017), wind-speed modeling (Nagarajan et al, 2015), and air quality monitoring (Delle Monache et al, 2014). Historically, analogue methods have been empirical, somewhat ad-hoc tools, but recently there have been attempts at putting these into a probabilistic framework (McDermott and Wikle, 2016;McDermott et al, 2018).…”
Section: Introductionmentioning
confidence: 99%
“…Forecasting is widely used for a wide range of scholarship and purposes, such as for data calculations [7], weather forecasting [8] and also used in foreign exchange [9].…”
Section: Forecastingmentioning
confidence: 99%
“…In the mallard settling pattern application, the temporal embedding lag of the input process is m = one month and we used q = 50 lagged months. Both values were chosen based on the priors used by McDermott et al (2018).…”
Section: D-eesn With Glmmentioning
confidence: 99%
“…The count data within transect is reported per 29km segment. In this application, we considered BPS count data for mallards (Anas platyrhynchos) from 1925 segments with a spatial extent of 85 − 165°W and 43 − 69°N for the years 1970 to 2014 obtained from (https://migbirdapps.fws.gov) to compare to previous works (Wu et al, 2013;McDermott et al, 2018).…”
Section: Background and Datamentioning
confidence: 99%