Search Results
. Chu and Zhao (2007) used the Bayesian probabilistic forecast models to predict the seasonal tropical cyclone activity in the central North Pacific. The Bayesian probabilistic models have also been used to predict the tropical cyclone activity in the North Atlantic ( Elsner and Jagger 2004 , 2006 ). Given the advantage of the probabilistic approach, we will adopt the Bayesian regression method for predicting the seasonal TC activity near Taiwan. As the objective of the present study is to develop
. Chu and Zhao (2007) used the Bayesian probabilistic forecast models to predict the seasonal tropical cyclone activity in the central North Pacific. The Bayesian probabilistic models have also been used to predict the tropical cyclone activity in the North Atlantic ( Elsner and Jagger 2004 , 2006 ). Given the advantage of the probabilistic approach, we will adopt the Bayesian regression method for predicting the seasonal TC activity near Taiwan. As the objective of the present study is to develop
at their base would, of course, heighten this volume effect ( Clark and Pollard 1998 ). In the following we explore the origins of changes in sea level sensitivity using a Bayesian methodology that combines observations of CO 2 and sea level with a physical model relating the two. Bayesian methods are well suited for inferring past climate conditions from uncertain proxy data sources ( Tingley and Huybers 2010 ) and have also been successfully applied to selecting between competing models for
at their base would, of course, heighten this volume effect ( Clark and Pollard 1998 ). In the following we explore the origins of changes in sea level sensitivity using a Bayesian methodology that combines observations of CO 2 and sea level with a physical model relating the two. Bayesian methods are well suited for inferring past climate conditions from uncertain proxy data sources ( Tingley and Huybers 2010 ) and have also been successfully applied to selecting between competing models for
of the likelihoods is 4 for low thresholds and 2 for higher thresholds (4 mm h −1 ). It is clearly seen that this approach produces a better discriminating ability between observed and nonobserved events in the likelihoods than the raw Bayesian model described above. The problem when this method is used for prognosis is that the actual value of the parameter θ must be estimated, since it is unknown in the moment that the forecast is being issued. As θ and, consequently, Δ are unknown at this
of the likelihoods is 4 for low thresholds and 2 for higher thresholds (4 mm h −1 ). It is clearly seen that this approach produces a better discriminating ability between observed and nonobserved events in the likelihoods than the raw Bayesian model described above. The problem when this method is used for prognosis is that the actual value of the parameter θ must be estimated, since it is unknown in the moment that the forecast is being issued. As θ and, consequently, Δ are unknown at this
-based Bayesian model averaging (En-BMA) approach, which is initially proposed for streamflow forecasting ( Darbandsari and Coulibaly 2020a ), an optimal subset of forecasts ensemble with lower dependency and higher information content was selected before applying the BMA method. Since detailed descriptions of the En-BMA concepts are provided in Darbandsari and Coulibaly (2020a) , a brief overview of this approach and the proposed modifications to make it more suitable for precipitation forecasting are
-based Bayesian model averaging (En-BMA) approach, which is initially proposed for streamflow forecasting ( Darbandsari and Coulibaly 2020a ), an optimal subset of forecasts ensemble with lower dependency and higher information content was selected before applying the BMA method. Since detailed descriptions of the En-BMA concepts are provided in Darbandsari and Coulibaly (2020a) , a brief overview of this approach and the proposed modifications to make it more suitable for precipitation forecasting are
predicted variable. When this “omitted-variable bias” is suspected to be large, care must be taken in interpreting the results. With these considerations in mind, we develop a Bayesian hierarchical model to estimate tornado reporting rates (TRRs) and actual expected tornado counts from the SPC tornado database. The method incorporates elements of previously published Bayesian approaches to this problem but adds a novel feature that addresses a serious solution nonuniqueness issue that may have impacted
predicted variable. When this “omitted-variable bias” is suspected to be large, care must be taken in interpreting the results. With these considerations in mind, we develop a Bayesian hierarchical model to estimate tornado reporting rates (TRRs) and actual expected tornado counts from the SPC tornado database. The method incorporates elements of previously published Bayesian approaches to this problem but adds a novel feature that addresses a serious solution nonuniqueness issue that may have impacted
weighted equally ( Räisänen and Palmer 2001 ). In many cases, combining ensemble results through Bayesian methods or weighted averages, where weights are determined by comparing model predictions to observations, shows improved performance better than simple averages (e.g., Barnston et al. 2003 ; Krishnamurti et al. 1999 , 2000 ; Robertson et al. 2004 ). The article by Tebaldi and Knutti (2007) presents a comprehensive literature review on the existing published methods to obtain best estimates
weighted equally ( Räisänen and Palmer 2001 ). In many cases, combining ensemble results through Bayesian methods or weighted averages, where weights are determined by comparing model predictions to observations, shows improved performance better than simple averages (e.g., Barnston et al. 2003 ; Krishnamurti et al. 1999 , 2000 ; Robertson et al. 2004 ). The article by Tebaldi and Knutti (2007) presents a comprehensive literature review on the existing published methods to obtain best estimates
observables exceeds that of the parameters, a dimensionally reducing version of the transformation-of-variables PDF conversion formula may be used ( Mardia et al. 1979 ; Lewis 2013b ), provided the observables can first be whitened (made independent and of equal variance, as in optimal fingerprint methods; Hegerl et al. 1996 ). The primary aims of this paper are to provide insight into the use of objective-Bayesian methods for estimating climate sensitivity by considering their relationship to
observables exceeds that of the parameters, a dimensionally reducing version of the transformation-of-variables PDF conversion formula may be used ( Mardia et al. 1979 ; Lewis 2013b ), provided the observables can first be whitened (made independent and of equal variance, as in optimal fingerprint methods; Hegerl et al. 1996 ). The primary aims of this paper are to provide insight into the use of objective-Bayesian methods for estimating climate sensitivity by considering their relationship to
informative prior estimation (IPE) for setting appropriate prior of Poisson rates, followed by framing it into the Bayesian model competing analysis. As a result, they built a more advanced framework for detecting and quantifying multiple (finite) shifts in a series of hurricane counts. b. Why RJMCMC? Although the MCMC imbedded with IPE method ( Zhao and Chu 2006 ) has been shown to be viable for calculating the posterior probability for a multiple hypothesis model, it suffers from a shortcoming. That is
informative prior estimation (IPE) for setting appropriate prior of Poisson rates, followed by framing it into the Bayesian model competing analysis. As a result, they built a more advanced framework for detecting and quantifying multiple (finite) shifts in a series of hurricane counts. b. Why RJMCMC? Although the MCMC imbedded with IPE method ( Zhao and Chu 2006 ) has been shown to be viable for calculating the posterior probability for a multiple hypothesis model, it suffers from a shortcoming. That is
,we apply a number of techniques, including the mixture model approach rather than the classical model posterior probability approach, the use of a prior of BMA weights that gives a slight preference toward more evenly distributed weights, and the use of the cross-validation likelihood function rather than the classical likelihood function in Bayesian inference. We conduct extensive analyses and find that our BMA method yields more stable weights and better forecast skill in cross validation than the
,we apply a number of techniques, including the mixture model approach rather than the classical model posterior probability approach, the use of a prior of BMA weights that gives a slight preference toward more evenly distributed weights, and the use of the cross-validation likelihood function rather than the classical likelihood function in Bayesian inference. We conduct extensive analyses and find that our BMA method yields more stable weights and better forecast skill in cross validation than the
estimation technique ( Silverman 1986 ) in Tan and Hong (2010) . They further assumed that the tornado occurrence follows a Poisson process to map tornado-induced wind velocity hazard. For the same region, Tan (2008) adopted the zero-inflated Poisson (ZIP) process with a Bayesian hierarchical modeling technique advocated by Wikle and Anderson (2003) to model the tornado occurrence and applied the Markov chain Monte Carlo (MCMC) method ( Gilks et al. 1996 ; Brooks et al. 2011 ) to estimate the model
estimation technique ( Silverman 1986 ) in Tan and Hong (2010) . They further assumed that the tornado occurrence follows a Poisson process to map tornado-induced wind velocity hazard. For the same region, Tan (2008) adopted the zero-inflated Poisson (ZIP) process with a Bayesian hierarchical modeling technique advocated by Wikle and Anderson (2003) to model the tornado occurrence and applied the Markov chain Monte Carlo (MCMC) method ( Gilks et al. 1996 ; Brooks et al. 2011 ) to estimate the model
