Bayesian Multiscale Smoothing for Making Inferences About Features in Scatterplots

Erästö, Panu; Holmström, Lasse

doi:10.1198/106186005x59315

Cited by 60 publications

(56 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…different transfer-function models, different data-sets, different records, different proxies) (Birks 1998). More sophisticated smoothers such as SiZer (significance of zero crossing of the derivative) (Chaudhuri & Marron 1999;Holmström & Erästö 2002) and the related BSiZer (Erästö & Holmström 2005) have considerable potential in palaeoclimatology because they can assess which features seen in a range of smoothed data are statistically significant and thus may be environmentally significant (see Korhola et al 2000;Erästö & Holmström 2005Weckström et al 2006 for palaeoecological applications). Given the random walk simulations of pollen data presented by Blaauw et al (2010), there is an increasing need to use techniques such as SiZer or BSiZer to distinguish signal from random red noise in stratigraphical time-series.…”

Section: Presentation and Interpretation Of Cli-mate Reconstructionsmentioning

confidence: 99%

Strengths and Weaknesses of Quantitative Climate Reconstructions Based on Late-Quaternary Biological Proxies

Birks

2011

TOECOLJ

343

284

View full text Add to dashboard Cite

Abstract:The importance of reconstructing past environments quantitatively in palaeoecology is reviewed by showing that many ecological questions asked of palaeoecological data commonly involve the reconstructions of past environment. Three basic approaches to reconstructing past climate from palaeoecological data are outlined and discussed in terms of their assumptions, strengths, and weaknesses. These approaches are the indicator-species approach involving bioclimateenvelope modelling; the assemblage approach involving modern analogue techniques and response surfaces; and the multivariate calibration-function approach. Topics common to all approaches are reviewed -presentation and interpretation, evaluation and validation, comparison, and general limitations of climate reconstructions. Challenges and possible future developments are presented and the potential future role of quantitative climate reconstructions in palaeoecology is summarised.

show abstract

Section: Presentation and Interpretation Of Cli-mate Reconstructionsmentioning

confidence: 99%

Strengths and Weaknesses of Quantitative Climate Reconstructions Based on Late-Quaternary Biological Proxies

Birks

2011

TOECOLJ

343

284

View full text Add to dashboard Cite

show abstract

“…However, such a point-wise inference is bound to result in a large number false positives, and we therefore use simultaneous inference over all times t k and a fixed σ. The simultaneous inference technique applied is the method of highest pointwise probabilities (HPW), first described in Erästö and Holmström (2005). The HPW is a greedy algorithm that selects time points according to their descending order of marginal posterior probability until the joint posterior probability of correlations being positive or negative is at least α.…”

Section: Feature Credibility Analysis Using Bayesian Inferencementioning

confidence: 99%

“…Further discussion and extensions of this model can be found in Erästö and Holmström (2005) and Pasanen et al (2013). The model for the time series y is defined similarly.…”

Section: Experiments 31 An Artificial Examplementioning

confidence: 99%

“…We take ν x0 = ν y0 = 10, which makes the priors of σ 2 x and σ 2 y vague. As in Erästö and Holmström (2005), full Bayesian inference could be used by considering κ x and κ y as unknown parameters of the model. However, for computational efficiency we have here used an empirical Bayes approach and estimated κ x and κ y and the prior means of σ 2 x and σ 2 y from the data.…”

Section: Experiments 31 An Artificial Examplementioning

confidence: 99%

See 1 more Smart Citation

Scale space multiresolution correlation analysis for time series data

Pasanen

Holmström

2016

Comput Stat

Self Cite

View full text Add to dashboard Cite

show abstract

“…This method was extended by Erästö and Holmström [2005], who incorporated Bayesian methods for identifying multiscale features. These methods have found applications in the analysis of climate data [Divine et al, 2007;Godtliebsen et al, 2003;Korhola et al, 2000;Rohling and Pälike, 2005] such as the reconstructed past temperature record.…”

Section: Previous Applications Of Scale Space To Climate Datamentioning

confidence: 99%

Scale space methods for climate model analysis

2013

View full text Add to dashboard Cite

[1] In this paper, we introduce methods for evaluating climate model performance across spatial scales. These techniques are based on the "scale space" framework widely used in the image processing and computer vision communities. We discuss why the diffusion equation on the sphere provides a particularly attractive means of smoothing two-dimensional maps of global climate data. We establish that no structure is introduced into a map as an artifact of the smoothing procedure. This allows for the comparison of models and observations at multiple scales. As a test case for these methods, we compare the ability of high-and low-resolution versions of the Community Climate System Model (CCSM) to simulate the seasonal climatologies of surface air temperature (TAS), sea level pressure (PSL), and total precipitation rate (PR). For TAS, we find that the high-resolution model is better able to capture the boreal summer (JJA) climatological pattern at fine scales, although there is no such improvement in winter (DJF). We find the performances of the high-and low-resolution models to be similarly capable of capturing the summertime sea level pressure climatology at all scales. However, the high-resolution model PSL climatology is degraded for DJF, especially at larger scales. For both JJA and DJF precipitation climatologies, we find larger precipitation errors in the high-resolution model at the finest scales; however, performance at larger scales is improved.

show abstract

Bayesian Multiscale Smoothing for Making Inferences About Features in Scatterplots

Cited by 60 publications

References 7 publications

Strengths and Weaknesses of Quantitative Climate Reconstructions Based on Late-Quaternary Biological Proxies

Strengths and Weaknesses of Quantitative Climate Reconstructions Based on Late-Quaternary Biological Proxies

Scale space multiresolution correlation analysis for time series data

Scale space methods for climate model analysis

Contact Info

Product

Resources

About