A Litterman BVAR approach for production forecasting of technology industries

Hsu, Po–Hsuan; Wang, Chi-Hsiu; Shyu, Joseph Z.; Yu, Hsien‐Chung

doi:10.1016/s0040-1625(01)00142-1

Cited by 14 publications

(12 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Second, the LBVAR model has been utilized frequently in economic forecasting for GDP, consumption quantity, and unemployment. In a recent study by Hsu et al (2003), the LBVAR model also performs well in production forecasting for technology industries. We are motivated to search for other BVAR models to make better production prediction in a more efficient way.…”

Section: Literature Reviewmentioning

confidence: 93%

“…First, we observed that various time series models have been used to predict industrial productions (e.g., Hsu et al, 2003;Marchetti & Parigi, 2000;Simpson, Osborn, & Sensier, 2001;Tseng et al, 1999). Second, we looked for a Bayesian multivariate time series model that fits unsteady environments better than traditional frequency-based models, and found that the non-informative diffuse-prior Bayesian vector autoregression (NDBVAR) model has good features: its prior is flexible and its computation is efficient.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Production Forecasting of Taiwan's Technology Industrial Cluster: A Bayesian Autoregression Approach

et al. 2005

Self Cite

View full text Add to dashboard Cite

This study proposes a forecasting method that combines the clustering effect and non‐informative diffuse‐prior Bayesian vector autoregression (NDBVAR) model to forecast the productions of technology industries. Two empirical cases are examined to verify the proposed method: the semiconductor industry and computer manufacturing industry in Taiwan. It is found that the NDBVAR model outperforms the other three conventional time series models including the autoregression (AR), vector autoregression (VAR), and Litterman Bayesian VAR (LBVAR) models. Moreover, the NDBVAR model also outperforms the forecast reports from leading market information providers over the past several years. The forecasting method proposed is therefore concluded to be a feasible approach for production prediction, especially for technology industries in volatile environments. Résumé La présente étude propose une méthode prévisionnelle qui combine les effets de regroupement et le non‐informative diffuse‐prior Bayesian vector autoregression model (NDBVAR) pour prévoir les productions des industries de technologie. Pour évaluer la méthode proposée, l'étude examine deux cas empiriques: les industries taiwanaises du semiconducteur et de fabrication d'ordinateur. Elle révèle que le modèle NDBVAR est plus performant que les trois modèles conventionnels en série chronologique notamment le modèle d'autoregression (AR), le modèle de vecteur d'autoregression (VAR), et le modèle Litterman Bayesian (LBVAR). L'étude montre aussi qu'au cours des dernières années, les modèles NDBVAR ont été plus performants que les rapports prévisionnels des prestataires d'informations qui dominent le marché. Elle débouche sur la constatation que la méthode prévisionnelle proposée est une approche réalisable pour la prévision de la production, en particulier pour les industries de la technologie dans un environnement volatile.

show abstract

Section: Literature Reviewmentioning

confidence: 93%

mentioning

confidence: 99%

Production Forecasting of Taiwan's Technology Industrial Cluster: A Bayesian Autoregression Approach

et al. 2005

Self Cite

View full text Add to dashboard Cite

show abstract

“…The system plots in Figure. Table 1.For the evaluation of approximation precision, the system adopts the weighted combination method [20] . The forecasts of the combination model are shown in Table 1.…”

Section: Models Combination With Wavelet Networkmentioning

confidence: 99%

Intelligent early-warning system of dam safety

Zhang²

2005

2005 International Conference on Machine Learning and Cybernetics

View full text Add to dashboard Cite

For the effective control of dam life process, it is important to implement the real-time diagnosis and reasonable evaluation for dam safety based on prototype observations. The development of intelligent early-warning systems of dam safety (IEWSDS) is a significant approach to realize above aim. This paper regarded dam as a vital and intelligent system, thereby constructed a bionics model of dam safety, which consisted of sensing system (nerve), central processing unit (cerebrum) and decision-making implement (organism). With above model and systems engineering, this paper designed an IEWSDS, which was composed of integration control, intelligent inference machine, project database, model base, graphics base and input/output modules. In the system, metadata base based on object oriented data model was used to implement the integration of homogeneous or nonhomogeneous databases storing dam safety data from different information sources. The multilevel link style was adopted to manage model base. Intelligent inference machine is the central processing unit of IEWSDS implementing data analysis, abnormity diagnosis and evaluation for dam safety. Because of strong non-linear and dynamic characteristics, the system adopted a combination model based on wavelet networks to approximate and forecast the run characteristics of dam. The abnormity diagnosis is a process of discovering knowledge on the pattern between of disease and geneses. The methods of attributions reduction in rough sets theory were presented to diagnose adaptively the abnormal cases and find out the geneses. The safety status of dam changes dynamically, which emerges in the quantitative and qualitative change manners. Therefore, quantitative and qualitative changes need be considered comprehensively in the process of dam safety evaluation. This paper proposed an extension evaluation method. The quantitative change and qualitative change of dam safety property were integrated into a matter-element. Based on the matter-element extensibility, the contradictory problem for dam safety elevation can be solved. In practice, the proposed system has been used to monitor dam safety successfully. The applications show that the bionics model is feasible, the proposed key technologies are effective. The systems can supply technical support for improving the level of dam safety management, extending normal run time of dam and voiding dam failure.

show abstract

“…Generally speaking, Time Serial Analysis performs better than other models as for the forecasting models with trendy and seasonal features. Abdel [2] and Hsu et al [3] adopted some Time Serials Models as forecasting models and found positive results. Recently with the development of the Artificial Intelligence Models, several methods are found to have better effectiveness than traditional models when being applied to forecasting models, and Artificial Neural Network (ANN) is the most commonly used tool applied in forecasting.…”

Section: Introductionmentioning

confidence: 99%

Re-engineering the forecasting phase using traditional and soft computing methods

Bertolini

Ciarapica

2010

2010 IEEE International Conference on Industrial Engineering and Engineering Management

View full text Add to dashboard Cite

The aim of the work is verifying the possibility of extrapolating information on demand trends, for a company specialized in the production of aluminium tins, using the data collected in previous periods. This study is mainly divided into three stages: (1) data pre-processing (data collection) stage, (2) adaptive network evaluating stage and (3) forecast and recall stage. At the stage of data collection, the data are divided into four categories: time serial data, macroeconomic data, downstream production demand data and industrial production data. The company analysed in this work usually carried out the prediction activities by means of expert judgement. In the case analyzed, four models were developed in order to predict the monthly number of tins: three traditional methods based on historical series and neural networks. Soft computing models were compared with traditional prediction models. Particularly the Holt-Winters forecasting method was tested developing a model that take into account seasonal phenomena.

show abstract

A Litterman BVAR approach for production forecasting of technology industries

Cited by 14 publications

References 28 publications

Production Forecasting of Taiwan's Technology Industrial Cluster: A Bayesian Autoregression Approach

Production Forecasting of Taiwan's Technology Industrial Cluster: A Bayesian Autoregression Approach

Intelligent early-warning system of dam safety

Re-engineering the forecasting phase using traditional and soft computing methods

Contact Info

Product

Resources

About