A bivariate first-order autoregressive time series model in exponential variables (Bear 1)
Article Abstract:
A non-complex time series model for bivariate exponential variables having a first-order autoregressive structure is developed called the BEAR (1) model. The New Exponential Autoregressive model (NEAR 2) is adapted to create the linear random coefficient difference equation model. This process is Markovian in the bivariate sense and has correlation development that is similar to that of the Gaussian AR(1) bivariate time series model. Research results indicate that the model shows a full range of cross-correlations and positive correlations. Research results also indicate that the marginal processes have correlation structure of ARMA (2,1) models.
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1989
User Contributions:
Comment about this article or add new information about this topic:
The parameter iteration method in dynamic programming
Article Abstract:
A practical method called the Parameter Iteration Method is developed for obtaining approximate solutions for problems involving systems with many components. Such multi-component systems would normally lead to impractical dynamic programming problems with too many state variables. An iteration method is used to overcome the computational difficulties and provide successive approximations of the value functions using recursive estimation and simulation. A problem of optimal replacement policy in a multi-item Markovian system is used as an example.
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1989
User Contributions:
Comment about this article or add new information about this topic:
An autoregressive process for beta random variables
Article Abstract:
Autoregressive processes involving stationary first-order autregression with beta marginal distribution are modeled and investigated. The models differ in that one has positive autocorrelation functions, whereas the other has alternating sign autocorrelation; both models exhibit linear and additive processes and both have beta random variables. The applications of these models to simulation techniques are discussed, as are the bivariate beta distributions of two consecutive observations and bivariate uniform processes.
Publication Name: Management Science
Subject: Business, general
ISSN: 0025-1909
Year: 1985
User Contributions:
Comment about this article or add new information about this topic:
- Abstracts: Dynamics of aggregate response sensitivities to marketing mix variables. From international to integrated marketing
- Abstracts: Industry and regional patterns in sequential foreign market entry. Structure and Performance in International Technology Transfer
- Abstracts: A comparison of futures pricing models in a new market: the case of individual share futures. Pricing Commodities When Both Price and Output Are Uncertain
- Abstracts: A Simulation Analysis of Alternative Pricing Strategies for Dynamic Environments. Markup variability and flexibility: theory and empirical evidence
- Abstracts: Interstate banking and small business finance: implications from available evidence. Can your small company acquire resources as favorably as the large company?