Verification in the presence of observation errors: Bayesian point of view
| Publication Date: |
2018-03-09
|
|---|---|
| Publisher: |
Wiley-Blackwell
|
| Print ISSN: |
0035-9009
|
| Electronic ISSN: |
1477-870X
|
| Topics: |
Geography
Physics
|
| Published by: |
| _version_ | 1839207948172132353 |
|---|---|
| autor | L. Duc, K. Saito |
| beschreibung | Verification in the presence of observation errors is approached form the Bayesian point of view. Like data assimilation (DA), Bayesian verification is shown to have a robust foundation established by Bayesian inference. Together, DA and Bayesian verification form two difference levels of Bayesian inference. Evaluation of a model is equivalent to inference on the plausibility of this model given observations. Relative performances between different models are measured by ratios of posterior plausibilities, which becomes ratios of likelihoods in case of no prior information. These ratios are called the Bayes factors and are the standard verification method in Bayesian model comparison. Since verification scores are used intensively in numerical weather prediction, the verification scores derived from likelihoods are proposed to replace the Bayes factors in Bayesian verification. With two requirements that the verification scores are both strictly proper and local, the logarithm score, i.e. log-likelihood, and its linear transformation are shown to be the unique class. Log-likelihoods in Bayesian verification are determined by the form of forecast probability distributions from models. The empirical form is preferable since its flexibility in incorporating not only observation errors but also other uncertainties in observation biases or observation error variances into calculation to obtain closed forms for log-likelihoods. When applied for observations with Gaussian errors, the logarithm score induces the weighted mean squared error which is non-dimensional and can be used for both univariate and multivariate observations. The most interesting application of Bayesian verification is to offer a new explanation for rank histograms and quantify the flatness of rank histograms by a metric which turns out to be the Kullback-Leibler divergence between the rank distribution observed in reality and a uniform rank distribution. It is worthy of note that the two very different metrics come from the logarithm score. |
| citation_standardnr | 6200625 |
| datenlieferant | ipn_articles |
| feed_id | 29506 |
| feed_publisher | Wiley-Blackwell |
| feed_publisher_url | http://www.wiley.com/wiley-blackwell |
| insertion_date | 2018-03-09 |
| journaleissn | 1477-870X |
| journalissn | 0035-9009 |
| publikationsjahr_anzeige | 2018 |
| publikationsjahr_facette | 2018 |
| publikationsjahr_intervall | 7984:2015-2019 |
| publikationsjahr_sort | 2018 |
| publisher | Wiley-Blackwell |
| quelle | Quarterly Journal of the Royal Meteorological Society |
| relation | http://onlinelibrary.wiley.com/resolve/doi?DOI=10.1002%2Fqj.3275 |
| search_space | articles |
| shingle_author_1 | L. Duc, K. Saito |
| shingle_author_2 | L. Duc, K. Saito |
| shingle_author_3 | L. Duc, K. Saito |
| shingle_author_4 | L. Duc, K. Saito |
| shingle_catch_all_1 | Verification in the presence of observation errors: Bayesian point of view Verification in the presence of observation errors is approached form the Bayesian point of view. Like data assimilation (DA), Bayesian verification is shown to have a robust foundation established by Bayesian inference. Together, DA and Bayesian verification form two difference levels of Bayesian inference. Evaluation of a model is equivalent to inference on the plausibility of this model given observations. Relative performances between different models are measured by ratios of posterior plausibilities, which becomes ratios of likelihoods in case of no prior information. These ratios are called the Bayes factors and are the standard verification method in Bayesian model comparison. Since verification scores are used intensively in numerical weather prediction, the verification scores derived from likelihoods are proposed to replace the Bayes factors in Bayesian verification. With two requirements that the verification scores are both strictly proper and local, the logarithm score, i.e. log-likelihood, and its linear transformation are shown to be the unique class. Log-likelihoods in Bayesian verification are determined by the form of forecast probability distributions from models. The empirical form is preferable since its flexibility in incorporating not only observation errors but also other uncertainties in observation biases or observation error variances into calculation to obtain closed forms for log-likelihoods. When applied for observations with Gaussian errors, the logarithm score induces the weighted mean squared error which is non-dimensional and can be used for both univariate and multivariate observations. The most interesting application of Bayesian verification is to offer a new explanation for rank histograms and quantify the flatness of rank histograms by a metric which turns out to be the Kullback-Leibler divergence between the rank distribution observed in reality and a uniform rank distribution. It is worthy of note that the two very different metrics come from the logarithm score. L. Duc, K. Saito Wiley-Blackwell 0035-9009 00359009 1477-870X 1477870X |
| shingle_catch_all_2 | Verification in the presence of observation errors: Bayesian point of view Verification in the presence of observation errors is approached form the Bayesian point of view. Like data assimilation (DA), Bayesian verification is shown to have a robust foundation established by Bayesian inference. Together, DA and Bayesian verification form two difference levels of Bayesian inference. Evaluation of a model is equivalent to inference on the plausibility of this model given observations. Relative performances between different models are measured by ratios of posterior plausibilities, which becomes ratios of likelihoods in case of no prior information. These ratios are called the Bayes factors and are the standard verification method in Bayesian model comparison. Since verification scores are used intensively in numerical weather prediction, the verification scores derived from likelihoods are proposed to replace the Bayes factors in Bayesian verification. With two requirements that the verification scores are both strictly proper and local, the logarithm score, i.e. log-likelihood, and its linear transformation are shown to be the unique class. Log-likelihoods in Bayesian verification are determined by the form of forecast probability distributions from models. The empirical form is preferable since its flexibility in incorporating not only observation errors but also other uncertainties in observation biases or observation error variances into calculation to obtain closed forms for log-likelihoods. When applied for observations with Gaussian errors, the logarithm score induces the weighted mean squared error which is non-dimensional and can be used for both univariate and multivariate observations. The most interesting application of Bayesian verification is to offer a new explanation for rank histograms and quantify the flatness of rank histograms by a metric which turns out to be the Kullback-Leibler divergence between the rank distribution observed in reality and a uniform rank distribution. It is worthy of note that the two very different metrics come from the logarithm score. L. Duc, K. Saito Wiley-Blackwell 0035-9009 00359009 1477-870X 1477870X |
| shingle_catch_all_3 | Verification in the presence of observation errors: Bayesian point of view Verification in the presence of observation errors is approached form the Bayesian point of view. Like data assimilation (DA), Bayesian verification is shown to have a robust foundation established by Bayesian inference. Together, DA and Bayesian verification form two difference levels of Bayesian inference. Evaluation of a model is equivalent to inference on the plausibility of this model given observations. Relative performances between different models are measured by ratios of posterior plausibilities, which becomes ratios of likelihoods in case of no prior information. These ratios are called the Bayes factors and are the standard verification method in Bayesian model comparison. Since verification scores are used intensively in numerical weather prediction, the verification scores derived from likelihoods are proposed to replace the Bayes factors in Bayesian verification. With two requirements that the verification scores are both strictly proper and local, the logarithm score, i.e. log-likelihood, and its linear transformation are shown to be the unique class. Log-likelihoods in Bayesian verification are determined by the form of forecast probability distributions from models. The empirical form is preferable since its flexibility in incorporating not only observation errors but also other uncertainties in observation biases or observation error variances into calculation to obtain closed forms for log-likelihoods. When applied for observations with Gaussian errors, the logarithm score induces the weighted mean squared error which is non-dimensional and can be used for both univariate and multivariate observations. The most interesting application of Bayesian verification is to offer a new explanation for rank histograms and quantify the flatness of rank histograms by a metric which turns out to be the Kullback-Leibler divergence between the rank distribution observed in reality and a uniform rank distribution. It is worthy of note that the two very different metrics come from the logarithm score. L. Duc, K. Saito Wiley-Blackwell 0035-9009 00359009 1477-870X 1477870X |
| shingle_catch_all_4 | Verification in the presence of observation errors: Bayesian point of view Verification in the presence of observation errors is approached form the Bayesian point of view. Like data assimilation (DA), Bayesian verification is shown to have a robust foundation established by Bayesian inference. Together, DA and Bayesian verification form two difference levels of Bayesian inference. Evaluation of a model is equivalent to inference on the plausibility of this model given observations. Relative performances between different models are measured by ratios of posterior plausibilities, which becomes ratios of likelihoods in case of no prior information. These ratios are called the Bayes factors and are the standard verification method in Bayesian model comparison. Since verification scores are used intensively in numerical weather prediction, the verification scores derived from likelihoods are proposed to replace the Bayes factors in Bayesian verification. With two requirements that the verification scores are both strictly proper and local, the logarithm score, i.e. log-likelihood, and its linear transformation are shown to be the unique class. Log-likelihoods in Bayesian verification are determined by the form of forecast probability distributions from models. The empirical form is preferable since its flexibility in incorporating not only observation errors but also other uncertainties in observation biases or observation error variances into calculation to obtain closed forms for log-likelihoods. When applied for observations with Gaussian errors, the logarithm score induces the weighted mean squared error which is non-dimensional and can be used for both univariate and multivariate observations. The most interesting application of Bayesian verification is to offer a new explanation for rank histograms and quantify the flatness of rank histograms by a metric which turns out to be the Kullback-Leibler divergence between the rank distribution observed in reality and a uniform rank distribution. It is worthy of note that the two very different metrics come from the logarithm score. L. Duc, K. Saito Wiley-Blackwell 0035-9009 00359009 1477-870X 1477870X |
| shingle_title_1 | Verification in the presence of observation errors: Bayesian point of view |
| shingle_title_2 | Verification in the presence of observation errors: Bayesian point of view |
| shingle_title_3 | Verification in the presence of observation errors: Bayesian point of view |
| shingle_title_4 | Verification in the presence of observation errors: Bayesian point of view |
| timestamp | 2025-07-31T23:43:02.177Z |
| titel | Verification in the presence of observation errors: Bayesian point of view |
| titel_suche | Verification in the presence of observation errors: Bayesian point of view |
| topic | R U |
| uid | ipn_articles_6200625 |