• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• 제55권3호
• /
• pp.206-216
• /
• 2019

This study is aimed to compare stock assessment models depending on how the models fit to observed data. Process-error model, Observation-error model, and Bayesian state-space model for the Korean Western coast fisheries were applied for comparison. Analytical results show that there is the least error between the estimated CPUE and the observed CPUE with the Bayesian state-space model; consequently, results of the Bayesian state-space model are the most reliable. According to the Bayesian State-space model, potential yield of fishery resources in the West Sea of Korea is estimated to be 231,949 tons per year. However, the results show that the fishery resources of West Sea have been decreasing since 1967. In addition, the amounts of stock in 2013 are assessed to be only 36% of the stock biomass at MSY level. Therefore, policy efforts are needed to recover the fishery resources of West Sea of Korea.

• Communications for Statistical Applications and Methods
• /
• 제10권1호
• /
• pp.203-210
• /
• 2003

In a standard Metropolis-type Monte Carlo simulation, the proposal distribution cannot be easily adapted to "local dynamics" of the target distribution. To overcome some of these difficulties, Duane et al. (1987) introduced the method of hybrid Monte Carlo(HMC) which combines the basic idea of molecular dynamics and the Metropolis acceptance-rejection rule to produce Monte Carlo samples from a given target distribution. In this paper, using the HMC within Gibbs sampler, an asymptotical estimate of the smoothing mean and a general solution to state space modeling in Bayesian framework is obtaineds obtained.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• 제53권6호
• /
• pp.925-933
• /
• 2020

This study is aimed to compare stock assessment models which are effective in assessing red snow crab Chinonoecetes japonicus resources and to select and apply an effective stock assessment model in the future. In order to select an effective stock assessment model, a process-error model, observation-error model, and a Bayesian state-space model were estimated. Analytical results show that the least error is observed between the estimated CPUE (catch per unit effort) and the observed CPUE when using the Bayesian state-space model. For the Bayesian state-space model, the 95% credible interval(CI) ranges for the maximum sustainable yield (MSY), carrying capacity (K), catchability coefficient (q), and intrinsic growth (r) are estimated to be 10,420-47,200 tons, 185,200-444,800 tons, 3.81E-06-9.02E-06, and 0.14-0.66, respectively. The results show that the Bayesian state-space model was most reliable among models.

• Journal of Korean Institute of Industrial Engineers
• /
• 제24권4호
• /
• pp.485-492
• /
• 1998

Most forecasting models often fail to produce appropriate forecasts because we build a model based on the assumption of the data being generated from the only one stochastic process. However, in many real problems, the time series data are generated from one stochastic process for a while and then abruptly undergo certain structural changes. In this paper, we assume the basic underlying process is the simple state-space model with random level and deterministic drift but interrupted by three types of exogenous shocks: level shift, drift change, outlier. A Bayesian procedure to detect, estimate and adapt to the structural changes is developed and compared with simple, double and adaptive exponential smoothing using simulated data and the U.S. leading composite index.

• Journal of Industrial Technology
• /
• 제43권1호
• /
• pp.57-62
• /
• 2023

In this review, we introduce the non-parametric Bayesian filtering algorithm known as the point-mass filter (PMF) and discuss recent studies related to it. PMF realizes Bayesian filtering by placing a deterministic grid on the state space and calculating the probability density at each grid point. PMF is known for its robustness and high accuracy compared to other nonparametric Bayesian filtering algorithms due to its uniform sampling. However, a drawback of PMF is its inherently high computational complexity in the prediction phase. In this review, we aim to understand the principles of the PMF algorithm and the reasons for the high computational complexity, and summarize recent research efforts to overcome this challenge. We hope that this review contributes to encouraging the consideration of PMF applications for various systems.

• Ocean and Polar Research
• /
• 제42권1호
• /
• pp.63-76
• /
• 2020

This study sought to assess the stock status and analyze the management effects with regard to the Black scraper, which is one of the more commercially important species in Korea. The catch amounts of Black scraper have significantly decreased since 1991. In this analysis, a Bayesian state-space model was utilized to assess the biomass of the Black scraper given the limited data. Model results showed that MSY and B_MSY of Black scraper were estimated to be 26,587 tons and 365,200 tons, respectively. In addition, the current biomass level of the Black scraper was assessed to be only 2.1% (7,549 tons) of B_MSY. For this reason, the effects of a moratorium policy on the Black scraper were evaluated. The results showed that if such a moratorium policy was implemented, it would take at least 18-40 years to restore the biomass level of the Black scraper to B_MSY depending upon its growth rates.

• Ocean and Polar Research
• /
• 제41권3호
• /
• pp.183-191
• /
• 2019

This study sought to find which model is most appropriate for estimating potential yield in the East Sea, Republic of Korea. For comparison purposes, the Process-error model, ASPIC model, Maximum entropy model, Observation-error model, and Bayesian state-space model were applied using data from catch amounts and total efforts of the whole catchable fishes in the East Sea. Results showed that the Bayesian state-space model was estimated to be the most reliable among the models. Potential yield of catchable species was estimated to be 227,858 tons per year. In addition, it was analyzed that the amount of fishery resources in 2016 was about 63% of the biomass that enables a fish stock to deliver the maximum sustainable yield.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• 제54권5호
• /
• pp.769-781
• /
• 2021

Given data about the annual fishery yield of the common squid Todarodes pacificus, and the catch-per-unit-effort (CPUE) data from multiple fisheries from 2000-2018, we applied a Bayesian state - space assessment model for the squid population. One of our objectives was to do a stock assessment, simultaneously incorporating CPUE data from the following three fisheries, (i) large trawl, (ii) jigger, and (iii) large purse seine, which comprised on average a year about 65% of all fisheries, allowing possible correlations to be reflected. Other objectives were to consider both observation and process errors and to apply objective priors of parameters. The estimated annual exploitable biomass was in the range of 3.50×10⁵ to 1.22×10⁶ MT, the estimated intrinsic growth rate was 1.02, and the estimated carrying capacity was 1,151,259 MT. Comparison with available results from stock assessment of independently analyzed single fisheries revealed a large difference from the estimated values, suggesting that stock assessment based on multiple fisheries should be performed.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2004년도 ICCAS
• /
• pp.631-634
• /
• 2004

In this paper, the Bayesian recurrent neural network (BRNN) is proposed to predict time series data. Among the various traditional prediction methodologies, a neural network method is considered to be more effective in case of non-linear and non-stationary time series data. A neural network predictor requests proper learning strategy to adjust the network weights, and one need to prepare for non-linear and non-stationary evolution of network weights. The Bayesian neural network in this paper estimates not the single set of weights but the probability distributions of weights. In other words, we sets the weight vector as a state vector of state space method, and estimates its probability distributions in accordance with the Bayesian inference. This approach makes it possible to obtain more exact estimation of the weights. Moreover, in the aspect of network architecture, it is known that the recurrent feedback structure is superior to the feedforward structure for the problem of time series prediction. Therefore, the recurrent network with Bayesian inference, what we call BRNN, is expected to show higher performance than the normal neural network. To verify the performance of the proposed method, the time series data are numerically generated and a neural network predictor is applied on it. As a result, BRNN is proved to show better prediction result than common feedforward Bayesian neural network.

• Fisheries and Aquatic Sciences
• /
• 제24권4호
• /
• pp.139-152
• /
• 2021

The main purpose of this study is to fit catch-per-unit-effort (CPUE) data about Korea chub mackerel (Scomber japonicus) stock with a state-space production (SSP) model, and to provide stock assessment results. We chose a surplus production model for the chub mackerel data, namely annual yield and CPUE. Then we employed a state-space layer for a production model to consider two sources of variability arising from unmodelled factors (process error) and noise in the data (observation error). We implemented the model via script software ADMB-RE because it reduces the computational cost of high-dimensional integration and provides Markov Chain Monte Carlo sampling, which is required for Bayesian approaches. To stabilize the numerical optimization, we considered prior distributions for model parameters. Applying the SSP model to data collected from commercial fisheries from 1999 to 2017, we estimated model parameters and management references, as well as uncertainties for the estimates. We also applied various production models and showed parameter estimates and goodness of fit statistics to compare the model performance. This study presents two significant findings. First, we concluded that the stock has been overexploited in terms of harvest rate from 1999 to 2017. Second, we suggest a SSP model for the smallest goodness of fit statistics among several production models, especially for fitting CPUE data with fluctuations.