The purpose of this study is to compare a negative binomial distribution with a negative binomial—Lindley by using stochastic orders. We characterize the comparisons in usual stochastic order, likelihood ratio order,...The purpose of this study is to compare a negative binomial distribution with a negative binomial—Lindley by using stochastic orders. We characterize the comparisons in usual stochastic order, likelihood ratio order, convex order, expectation order and uniformly more variable order based on theorem and some numerical example of comparisons between negative binomial random variable and negative binomial—Lindley random variable.展开更多
In the reputation modeling of wireless sensor networks(WSNs) many literatures have proposed creative reputation indirect update methods,such as reputation integration,discounting,aging to eliminate,and filtering mal...In the reputation modeling of wireless sensor networks(WSNs) many literatures have proposed creative reputation indirect update methods,such as reputation integration,discounting,aging to eliminate,and filtering malicious reputation information. However,few have discussed the reputation direct update. In this paper,based on sound statistical theories,a negative binominal distribution method in the reputation direct update for WSNs is proposed. Results show that the proposed method is more suitable and time-saving for the reputation update of the resource constraint WSNs and can improve the computation power efficiency as well.展开更多
In this article, the zero-inflated non-central negative binomial(ZINNB) distribution is introduced. Some of its basic properties are obtained. In addition, we use the maximum likelihood estimation method to estimate t...In this article, the zero-inflated non-central negative binomial(ZINNB) distribution is introduced. Some of its basic properties are obtained. In addition, we use the maximum likelihood estimation method to estimate the parameters of the ZINNB distribution, and illustrate its application by fitting the actual data sets.展开更多
Zero-inflated negative binomial distribution is characterized in this paper through a linear differential equation satisfied by its probability generating function.
This paper discusses a characterization of the members of a subfamily of power series distributions when their probability generating functions satisfy the functional equation where a, b and c are constants and is the...This paper discusses a characterization of the members of a subfamily of power series distributions when their probability generating functions satisfy the functional equation where a, b and c are constants and is the derivative of f.展开更多
In this paper we extend and improve some results of the large deviation for random sums of random variables. Let {Xn;n 〉 1} be a sequence of non-negative, independent and identically distributed random variables with...In this paper we extend and improve some results of the large deviation for random sums of random variables. Let {Xn;n 〉 1} be a sequence of non-negative, independent and identically distributed random variables with common heavy-tailed distribution function F and finite mean μ ∈R^+, {N(n); n ≥0} be a sequence of negative binomial distributed random variables with a parameter p C (0, 1), n ≥ 0, let {M(n); n ≥ 0} be a Poisson process with intensity λ 〉 0. Suppose {N(n); n ≥ 0}, {Xn; n≥1} and {M(n); n ≥ 0} are mutually independent. Write S(n) =N(n)∑i=1 Xi-cM(n).Under the assumption F ∈ C, we prove some large deviation results. These results can be applied to certain problems in insurance and finance.展开更多
In this paper,we study a robust estimation method for the observation-driven integervalued time-series models in which the conditional probability mass of current observations is assumed to follow a negative binomial ...In this paper,we study a robust estimation method for the observation-driven integervalued time-series models in which the conditional probability mass of current observations is assumed to follow a negative binomial distribution.Maximum likelihood estimator is highly affected by the outliers.We resort to the minimum density power divergence estimator as a robust estimator and showthat it is strongly consistent and asymptotically normal under some regularity conditions.Simulation results are provided to illustrate the performance of the estimator.An application is performed on data for campylobacteriosis infections.展开更多
The compound negative binomial model, introduced in this paper, is a discrete time version. We discuss the Markov properties of the surplus process, and study the ruin probability and the joint distributions of actuar...The compound negative binomial model, introduced in this paper, is a discrete time version. We discuss the Markov properties of the surplus process, and study the ruin probability and the joint distributions of actuarial random vectors in this model. By the strong Markov property and the mass function of a defective renewal sequence, we obtain the explicit expressions of the ruin probability, the finite-horizon ruin probability, the joint distributions of T, U(T - 1), |U(T)| and inf U(n) (i.e., the time of ruin, the surplus immediately before ruin, the deficit at ruin and maximal deficit from ruin to recovery) and the distributions of some actuarial random vectors.展开更多
The rharginal recursive equations on excess-of-loss reinsurance treaty are investignted, under the assumption that the number of claims belongs to the family consisting of Poisson, binomial and negative binomial, and ...The rharginal recursive equations on excess-of-loss reinsurance treaty are investignted, under the assumption that the number of claims belongs to the family consisting of Poisson, binomial and negative binomial, and that the severity distribution has bounded continuous density function. On conditional of the numbers of claims associated with the reinsurer and the cedent, some recursive equations are obtained for the marginal distributions of the total payments of the reinsurer and the cedent.展开更多
Background: In this paper, a regression model for predicting the spatial distribution of forest cockchafer larvae in the Hessian Ried region (Germany) is presented. The forest cockchafer, a native biotic pest, is a...Background: In this paper, a regression model for predicting the spatial distribution of forest cockchafer larvae in the Hessian Ried region (Germany) is presented. The forest cockchafer, a native biotic pest, is a major cause of damage in forests in this region particularly during the regeneration phase. The model developed in this study is based on a systematic sample inventory of forest cockchafer larvae by excavation across the Hessian Ried. These forest cockchafer larvae data were characterized by excess zeros and overdispersion. Methods: Using specific generalized additive regression models, different discrete distributions, including the Poisson, negative binomial and zero-inflated Poisson distributions, were compared. The methodology employed allowed the simultaneous estimation of non-linear model effects of causal covariates and, to account for spatial autocorrelation, of a 2-dimensional spatial trend function. In the validation of the models, both the Akaike information criterion (AIC) and more detailed graphical procedures based on randomized quantile residuals were used. Results: The negative binomial distribution was superior to the Poisson and the zero-inflated Poisson distributions, providing a near perfect fit to the data, which was proven in an extensive validation process. The causal predictors found to affect the density of larvae significantly were distance to water table and percentage of pure clay layer in the soil to a depth of I m. Model predictions showed that larva density increased with an increase in distance to the water table up to almost 4 m, after which it remained constant, and with a reduction in the percentage of pure clay layer. However this latter correlation was weak and requires further investigation. The 2-dimensional trend function indicated a strong spatial effect, and thus explained by far the highest proportion of variation in larva density. Conclusions: As such the model can be used to support forest practitioners in their decision making for regeneration and forest protection planning in the Hessian predicting future spatial patterns of the larva density is still comparatively weak. Ried. However, the application of the model for somewhat limited because the causal effects are展开更多
The World Wide Web is essential to general public nowadays. From a data analysis viewpoint, it provides rich opportunities to gather observational data on a large-scale. This paper focuses on modeling the behavior of ...The World Wide Web is essential to general public nowadays. From a data analysis viewpoint, it provides rich opportunities to gather observational data on a large-scale. This paper focuses on modeling the behavior of visitors to an academic website. Although the conventional probability models, which were used by other literature for fitting in a commercial web site, capture the power law behavior in our data, they fail to capture other important features like the long tail. We propose a new model based on the identities of the users. Qualitative and quantitative tests, which are used for comparing the model fitting to our data, show that the new model outperforms other two conventional probability models.展开更多
An attempt is made to study various types of multiplicity distributions and multiplicity correlations amongst secondary charged particles produced in 4.5 and 14.5 A GeV/c 28Si-nucleus interactions. The results reveal ...An attempt is made to study various types of multiplicity distributions and multiplicity correlations amongst secondary charged particles produced in 4.5 and 14.5 A GeV/c 28Si-nucleus interactions. The results reveal that multiplicity correlations of the type , where i, j = b, g, s and h with i ≠ j, are linear. The observed multiplicity correlations are nicely fitted by the method of least squares fitting of the type, = b + ani. Furthermore, the multiplicity distributions of relativistic charged particles and compound multiplicity at 4.5 A GeV/c are nicely fitted by Poisson distributions with peaks at relatively lower multiplicities. However, in the case of multiplicity distributions of various types of secondary charged particles produced in 14.5 A GeV/c 28Si-nucleus collisions definite trends are not discernible.展开更多
We investigate the major characteristics of the occurrences, causes of and counter measures for aircraft accidents in Japan. We apply statistical data analysis and mathematical modeling techniques to determine the rel...We investigate the major characteristics of the occurrences, causes of and counter measures for aircraft accidents in Japan. We apply statistical data analysis and mathematical modeling techniques to determine the relations among economic growth, aviation demand, the frequency of aircraft/helicopter accidents, the major characteristics of the occurrence intervals of accidents, and the number of fatalities due to accidents. The statistical model analysis suggests that the occurrence intervals of accidents and the number of fatalities can be explained by probability distributions such as the exponential distribution and the negative binomial distribution, respectively. We show that countermeasures for preventing accidents have been developed in every aircraft model, and thus they have contributed to a significant decrease in the number of accidents in the last three decades. We find that the major cause of accidents involving large airplanes has been weather, while accidents involving small airplanes and helicopters are mainly due to the pilot error. We also discover that, with respect to accidents mainly due to pilot error, there is a significant decrease in the number of accidents due to the aging of airplanes, whereas the number of accidents due to weather has barely declined. We further determine that accidents involving small and large airplanes mostly occur during takeoff and landing, whereas those involving helicopters are most likely to happen during flight. In order to decrease the number of accidents, i) enhancing safety and security by further developing technologies for aircraft, airports and air control radars, ii) establishing and improving training methods for crew including pilots, mechanics and traffic controllers, iii) tightening public rules, and iv) strengthening efforts made by individual aviation-related companies are absolutely necessary.展开更多
In this article we propose a novel hurdle negative binomial (HNB) regression combined with a distributed lag nonlinear model (DLNM) to model weather factors’ impact on heat related illness (HRI) in Singapore. AIC cri...In this article we propose a novel hurdle negative binomial (HNB) regression combined with a distributed lag nonlinear model (DLNM) to model weather factors’ impact on heat related illness (HRI) in Singapore. AIC criterion is adopted to help select proper combination of weather variables and check their lagged effect as well as nonlinear effect. The process of model selection and validation is demonstrated. It is observed that the predicted occurrence rate is close to the observed one. The proposed combined model can be used to predict HRI cases for mitigating HRI occurrences and provide inputs for related public health policy considering climate change impact.展开更多
In this paper an effort has been made to study the general characteristics of slow particles produced in the interactions of 32S-Em at 200 AGeV to extract the information about the mechanism of particle production. Th...In this paper an effort has been made to study the general characteristics of slow particles produced in the interactions of 32S-Em at 200 AGeV to extract the information about the mechanism of particle production. The results have been compared with the experimental results obtained by other workers. The multiplicity distributions of the slow target associated particles (black, grey and heavy tracks) produced by 32S-beam with different targets have been studied. Also several types of correlations among them have been investigated. The variation of the produced particles with projectile mass number and target size has been studied. Also the multiplicity distributions of slow particles with NBD fits are presented and scaling multiplicity distributions of slow particles produced have been studied in order to check the validity of KNO-scaling.展开更多
Reputation mechanism has been recently introduced into wireless sensor networks(WSNs) to overcome the deficiencies incurred by cryptography alone.Unlike most current reputation mechanisms that are based on binomial di...Reputation mechanism has been recently introduced into wireless sensor networks(WSNs) to overcome the deficiencies incurred by cryptography alone.Unlike most current reputation mechanisms that are based on binomial distribution and to some extent,do not give due attention to the aspect of energy constraint of WSNs,this research deduces and proves the feasibility that negative binomial distribution can well be used in the reputation update with distinctive energy-saving features.Comparison tests with respect to the energy consumption in terms of reputation computing frequencies are done between the traditional reputation method and the one in this study.Results show that our method can save more energy for the reputation update and thus can better meet the power constraints of WSNs.展开更多
This paper proposes a new customer lifetime model:the Gamma/Weibull distribution(G/W).Similar to the Pareto/NBD model,we propose a G/W/NBD model by combining the G/W distribution with a negative binomial distribution(...This paper proposes a new customer lifetime model:the Gamma/Weibull distribution(G/W).Similar to the Pareto/NBD model,we propose a G/W/NBD model by combining the G/W distribution with a negative binomial distribution(NBD)and study its properties such as(i)the probability that a customer to be alive at a time point;(ii)the expectation and variance of the number of transactions for a customer during a fixed time period;(iii)the conditional expectation and conditional variance of the number of future transactions for a customer during a fixed time period.Several simulation studies are conducted to investigate the forecasting accuracy and flexibility of the proposed model.A CDNOW data set is analyzed by the proposed model.展开更多
We consider a Markov chain X = {Xi, i = 1, 2,...} with the state space {0, 1}, and define W = ∑1=1^x XiXi+1, which is the number of 2-runs in X before time n + 1. In this paper, we prove that the negative binomial ...We consider a Markov chain X = {Xi, i = 1, 2,...} with the state space {0, 1}, and define W = ∑1=1^x XiXi+1, which is the number of 2-runs in X before time n + 1. In this paper, we prove that the negative binomial distribution is an appropriate approximation for LW when VarW is greater than EW. The error estimate obtained herein improves the corresponding result in previous literatures.展开更多
Aims Fits of species-abundance distributions to empirical data are increasingly used to evaluate models of diversity maintenance and community structure and to infer properties of communities,such as species richness....Aims Fits of species-abundance distributions to empirical data are increasingly used to evaluate models of diversity maintenance and community structure and to infer properties of communities,such as species richness.Two distributions predicted by several models are the Poisson lognormal(PLN)and the negative binomial(NB)distribution;however,at least three different ways to parameterize the PLN have been proposed,which differ in whether unobserved species contribute to the likelihood and in whether the likelihood is conditional upon the total number of individuals in the sample.Each of these has an analogue for the NB.Here,we propose a new formulation of the PLN and NB that includes the number of unobserved species as one of the estimated parameters.We investigate the performance of parameter estimates obtained from this reformulation,as well as the existing alternatives,for drawing inferences about the shape of species abundance distributions and estimation of species richness.Methods We simulate the random sampling of a fixed number of individuals from lognormal and gamma community relative abundance distributions,using a previously developed‘individual-based’bootstrap algorithm.We use a range of sample sizes,community species richness levels and shape parameters for the species abundance distributions that span much of the realistic range for empirical data,generating 1000 simulated data sets for each parameter combination.We then fit each of the alternative likelihoods to each of the simulated data sets,and we assess the bias,sampling variance and estimation error for each method.Important Findings Parameter estimates behave reasonably well for most parameter values,exhibiting modest levels of median error.However,for the NB,median error becomes extremely large as the NB approaches either of two limiting cases.For both the NB and PLN,>90%of the variation in the error in model parameters across parameter sets is explained by three quantities that corresponded to the proportion of species not observed in the sample,the expected number of species observed in the sample and the discrepancy between the true NB or PLN distribution and a Poisson distribution with the same mean.There are relatively few systematic differences between the four alternative likelihoods.In particular,failing to condition the likelihood on the total sample sizes does not appear to systematically increase the bias in parameter estimates.Indeed,overall,the classical likelihood performs slightly better than the alternatives.However,our reparameterized likelihood,for which species richness is a fitted parameter,has important advantages over existing approaches for estimating species richness from fitted species-abundance models.展开更多
文摘The purpose of this study is to compare a negative binomial distribution with a negative binomial—Lindley by using stochastic orders. We characterize the comparisons in usual stochastic order, likelihood ratio order, convex order, expectation order and uniformly more variable order based on theorem and some numerical example of comparisons between negative binomial random variable and negative binomial—Lindley random variable.
基金supported by the National Natural Science Foundation of China under Grant No. 6107311
文摘In the reputation modeling of wireless sensor networks(WSNs) many literatures have proposed creative reputation indirect update methods,such as reputation integration,discounting,aging to eliminate,and filtering malicious reputation information. However,few have discussed the reputation direct update. In this paper,based on sound statistical theories,a negative binominal distribution method in the reputation direct update for WSNs is proposed. Results show that the proposed method is more suitable and time-saving for the reputation update of the resource constraint WSNs and can improve the computation power efficiency as well.
文摘In this article, the zero-inflated non-central negative binomial(ZINNB) distribution is introduced. Some of its basic properties are obtained. In addition, we use the maximum likelihood estimation method to estimate the parameters of the ZINNB distribution, and illustrate its application by fitting the actual data sets.
文摘Zero-inflated negative binomial distribution is characterized in this paper through a linear differential equation satisfied by its probability generating function.
文摘This paper discusses a characterization of the members of a subfamily of power series distributions when their probability generating functions satisfy the functional equation where a, b and c are constants and is the derivative of f.
文摘In this paper we extend and improve some results of the large deviation for random sums of random variables. Let {Xn;n 〉 1} be a sequence of non-negative, independent and identically distributed random variables with common heavy-tailed distribution function F and finite mean μ ∈R^+, {N(n); n ≥0} be a sequence of negative binomial distributed random variables with a parameter p C (0, 1), n ≥ 0, let {M(n); n ≥ 0} be a Poisson process with intensity λ 〉 0. Suppose {N(n); n ≥ 0}, {Xn; n≥1} and {M(n); n ≥ 0} are mutually independent. Write S(n) =N(n)∑i=1 Xi-cM(n).Under the assumption F ∈ C, we prove some large deviation results. These results can be applied to certain problems in insurance and finance.
基金supported by National Natural Science Foundation of China(Nos.11871027,11731015)Science and Technology Developing Plan of Jilin Province(No.20170101057JC)Cultivation Plan for Excellent Young Scholar Candidates of Jilin University.
文摘In this paper,we study a robust estimation method for the observation-driven integervalued time-series models in which the conditional probability mass of current observations is assumed to follow a negative binomial distribution.Maximum likelihood estimator is highly affected by the outliers.We resort to the minimum density power divergence estimator as a robust estimator and showthat it is strongly consistent and asymptotically normal under some regularity conditions.Simulation results are provided to illustrate the performance of the estimator.An application is performed on data for campylobacteriosis infections.
基金Supported by the National Natural Science Foundation of China (No.10671197)
文摘The compound negative binomial model, introduced in this paper, is a discrete time version. We discuss the Markov properties of the surplus process, and study the ruin probability and the joint distributions of actuarial random vectors in this model. By the strong Markov property and the mass function of a defective renewal sequence, we obtain the explicit expressions of the ruin probability, the finite-horizon ruin probability, the joint distributions of T, U(T - 1), |U(T)| and inf U(n) (i.e., the time of ruin, the surplus immediately before ruin, the deficit at ruin and maximal deficit from ruin to recovery) and the distributions of some actuarial random vectors.
基金Project supported by the National Natural Science Foundation of China (Nos. 10471008, 19831020)
文摘The rharginal recursive equations on excess-of-loss reinsurance treaty are investignted, under the assumption that the number of claims belongs to the family consisting of Poisson, binomial and negative binomial, and that the severity distribution has bounded continuous density function. On conditional of the numbers of claims associated with the reinsurer and the cedent, some recursive equations are obtained for the marginal distributions of the total payments of the reinsurer and the cedent.
文摘Background: In this paper, a regression model for predicting the spatial distribution of forest cockchafer larvae in the Hessian Ried region (Germany) is presented. The forest cockchafer, a native biotic pest, is a major cause of damage in forests in this region particularly during the regeneration phase. The model developed in this study is based on a systematic sample inventory of forest cockchafer larvae by excavation across the Hessian Ried. These forest cockchafer larvae data were characterized by excess zeros and overdispersion. Methods: Using specific generalized additive regression models, different discrete distributions, including the Poisson, negative binomial and zero-inflated Poisson distributions, were compared. The methodology employed allowed the simultaneous estimation of non-linear model effects of causal covariates and, to account for spatial autocorrelation, of a 2-dimensional spatial trend function. In the validation of the models, both the Akaike information criterion (AIC) and more detailed graphical procedures based on randomized quantile residuals were used. Results: The negative binomial distribution was superior to the Poisson and the zero-inflated Poisson distributions, providing a near perfect fit to the data, which was proven in an extensive validation process. The causal predictors found to affect the density of larvae significantly were distance to water table and percentage of pure clay layer in the soil to a depth of I m. Model predictions showed that larva density increased with an increase in distance to the water table up to almost 4 m, after which it remained constant, and with a reduction in the percentage of pure clay layer. However this latter correlation was weak and requires further investigation. The 2-dimensional trend function indicated a strong spatial effect, and thus explained by far the highest proportion of variation in larva density. Conclusions: As such the model can be used to support forest practitioners in their decision making for regeneration and forest protection planning in the Hessian predicting future spatial patterns of the larva density is still comparatively weak. Ried. However, the application of the model for somewhat limited because the causal effects are
文摘The World Wide Web is essential to general public nowadays. From a data analysis viewpoint, it provides rich opportunities to gather observational data on a large-scale. This paper focuses on modeling the behavior of visitors to an academic website. Although the conventional probability models, which were used by other literature for fitting in a commercial web site, capture the power law behavior in our data, they fail to capture other important features like the long tail. We propose a new model based on the identities of the users. Qualitative and quantitative tests, which are used for comparing the model fitting to our data, show that the new model outperforms other two conventional probability models.
文摘An attempt is made to study various types of multiplicity distributions and multiplicity correlations amongst secondary charged particles produced in 4.5 and 14.5 A GeV/c 28Si-nucleus interactions. The results reveal that multiplicity correlations of the type , where i, j = b, g, s and h with i ≠ j, are linear. The observed multiplicity correlations are nicely fitted by the method of least squares fitting of the type, = b + ani. Furthermore, the multiplicity distributions of relativistic charged particles and compound multiplicity at 4.5 A GeV/c are nicely fitted by Poisson distributions with peaks at relatively lower multiplicities. However, in the case of multiplicity distributions of various types of secondary charged particles produced in 14.5 A GeV/c 28Si-nucleus collisions definite trends are not discernible.
文摘We investigate the major characteristics of the occurrences, causes of and counter measures for aircraft accidents in Japan. We apply statistical data analysis and mathematical modeling techniques to determine the relations among economic growth, aviation demand, the frequency of aircraft/helicopter accidents, the major characteristics of the occurrence intervals of accidents, and the number of fatalities due to accidents. The statistical model analysis suggests that the occurrence intervals of accidents and the number of fatalities can be explained by probability distributions such as the exponential distribution and the negative binomial distribution, respectively. We show that countermeasures for preventing accidents have been developed in every aircraft model, and thus they have contributed to a significant decrease in the number of accidents in the last three decades. We find that the major cause of accidents involving large airplanes has been weather, while accidents involving small airplanes and helicopters are mainly due to the pilot error. We also discover that, with respect to accidents mainly due to pilot error, there is a significant decrease in the number of accidents due to the aging of airplanes, whereas the number of accidents due to weather has barely declined. We further determine that accidents involving small and large airplanes mostly occur during takeoff and landing, whereas those involving helicopters are most likely to happen during flight. In order to decrease the number of accidents, i) enhancing safety and security by further developing technologies for aircraft, airports and air control radars, ii) establishing and improving training methods for crew including pilots, mechanics and traffic controllers, iii) tightening public rules, and iv) strengthening efforts made by individual aviation-related companies are absolutely necessary.
文摘In this article we propose a novel hurdle negative binomial (HNB) regression combined with a distributed lag nonlinear model (DLNM) to model weather factors’ impact on heat related illness (HRI) in Singapore. AIC criterion is adopted to help select proper combination of weather variables and check their lagged effect as well as nonlinear effect. The process of model selection and validation is demonstrated. It is observed that the predicted occurrence rate is close to the observed one. The proposed combined model can be used to predict HRI cases for mitigating HRI occurrences and provide inputs for related public health policy considering climate change impact.
文摘In this paper an effort has been made to study the general characteristics of slow particles produced in the interactions of 32S-Em at 200 AGeV to extract the information about the mechanism of particle production. The results have been compared with the experimental results obtained by other workers. The multiplicity distributions of the slow target associated particles (black, grey and heavy tracks) produced by 32S-beam with different targets have been studied. Also several types of correlations among them have been investigated. The variation of the produced particles with projectile mass number and target size has been studied. Also the multiplicity distributions of slow particles with NBD fits are presented and scaling multiplicity distributions of slow particles produced have been studied in order to check the validity of KNO-scaling.
基金the National Natural Science Foundation of China(No.60905037)
文摘Reputation mechanism has been recently introduced into wireless sensor networks(WSNs) to overcome the deficiencies incurred by cryptography alone.Unlike most current reputation mechanisms that are based on binomial distribution and to some extent,do not give due attention to the aspect of energy constraint of WSNs,this research deduces and proves the feasibility that negative binomial distribution can well be used in the reputation update with distinctive energy-saving features.Comparison tests with respect to the energy consumption in terms of reputation computing frequencies are done between the traditional reputation method and the one in this study.Results show that our method can save more energy for the reputation update and thus can better meet the power constraints of WSNs.
文摘This paper proposes a new customer lifetime model:the Gamma/Weibull distribution(G/W).Similar to the Pareto/NBD model,we propose a G/W/NBD model by combining the G/W distribution with a negative binomial distribution(NBD)and study its properties such as(i)the probability that a customer to be alive at a time point;(ii)the expectation and variance of the number of transactions for a customer during a fixed time period;(iii)the conditional expectation and conditional variance of the number of future transactions for a customer during a fixed time period.Several simulation studies are conducted to investigate the forecasting accuracy and flexibility of the proposed model.A CDNOW data set is analyzed by the proposed model.
基金Supported by National Natural Science Foundation of China(Grant No.11071021)
文摘We consider a Markov chain X = {Xi, i = 1, 2,...} with the state space {0, 1}, and define W = ∑1=1^x XiXi+1, which is the number of 2-runs in X before time n + 1. In this paper, we prove that the negative binomial distribution is an appropriate approximation for LW when VarW is greater than EW. The error estimate obtained herein improves the corresponding result in previous literatures.
基金Australian Research Council(grants CE0561435 and DP0880544)James Cook University.
文摘Aims Fits of species-abundance distributions to empirical data are increasingly used to evaluate models of diversity maintenance and community structure and to infer properties of communities,such as species richness.Two distributions predicted by several models are the Poisson lognormal(PLN)and the negative binomial(NB)distribution;however,at least three different ways to parameterize the PLN have been proposed,which differ in whether unobserved species contribute to the likelihood and in whether the likelihood is conditional upon the total number of individuals in the sample.Each of these has an analogue for the NB.Here,we propose a new formulation of the PLN and NB that includes the number of unobserved species as one of the estimated parameters.We investigate the performance of parameter estimates obtained from this reformulation,as well as the existing alternatives,for drawing inferences about the shape of species abundance distributions and estimation of species richness.Methods We simulate the random sampling of a fixed number of individuals from lognormal and gamma community relative abundance distributions,using a previously developed‘individual-based’bootstrap algorithm.We use a range of sample sizes,community species richness levels and shape parameters for the species abundance distributions that span much of the realistic range for empirical data,generating 1000 simulated data sets for each parameter combination.We then fit each of the alternative likelihoods to each of the simulated data sets,and we assess the bias,sampling variance and estimation error for each method.Important Findings Parameter estimates behave reasonably well for most parameter values,exhibiting modest levels of median error.However,for the NB,median error becomes extremely large as the NB approaches either of two limiting cases.For both the NB and PLN,>90%of the variation in the error in model parameters across parameter sets is explained by three quantities that corresponded to the proportion of species not observed in the sample,the expected number of species observed in the sample and the discrepancy between the true NB or PLN distribution and a Poisson distribution with the same mean.There are relatively few systematic differences between the four alternative likelihoods.In particular,failing to condition the likelihood on the total sample sizes does not appear to systematically increase the bias in parameter estimates.Indeed,overall,the classical likelihood performs slightly better than the alternatives.However,our reparameterized likelihood,for which species richness is a fitted parameter,has important advantages over existing approaches for estimating species richness from fitted species-abundance models.