Bayesian forecasting for seemingly unrelated time series: application to local government revenue forecasting

Article Abstract:

A comparison of two Bayesian forecasting methods is made through their application on an income tax revenue forecasting study in 40 Pennsylvania school districts. The Multi-State Kalman Filter (MSKF) method is used on time series data that are short, irregular and that come from parallel sources within the same environmental context. Because such data are subject tothe same external influences, combining the MSKF method with the Conditionally Independent Hierarchial method provides for the simultaneous treatment of such series. This modified technique, called the C-MSKF method, is shown to be more accurate than the univariate MSKF. The performance of C-MSKF is enhanced in thepresence of decreasing series lengths, increasing forecast horizons, and schooldistrict sensitivity to the business cycle.

Methods, Usage, Bayesian statistical decision theory, Bayesian analysis, Forecasting, Tax revenue estimating

---

A value efficiency approach to incorporating preference information in data envelopment analysis

Article Abstract:

Data Envelopment Analysis (DEA) is a useful tool for quantifying the performance of Decision Making Units (DMUs) that essentially have the same tasks and generate similar multiple outputs using similar multiple resources. This analytical tool identifies efficient and inefficient units and provides information on efficiency scores and reference sets for the latter. The efficiency scores and other DEA results can serve as performance indicators of DMUs. An operational procedure and a supporting theory needed to integrate the Decision Maker's (DM) preference information into DEA type efficiency analysis are presented. The DMs' preferences are modelled through their Most Preferred Solution. The model can be used immediately and easily to solve practical problems.

Decision-making, Models, Decision making, Analysis, Data envelopment analysis, Information theory

---

Similar abstracts:

• Abstracts: A role for the forecasting function. How good are neural networks for causal forecasting?
• Abstracts: Economic forecasting in a business environment. Taking the con out of economic forecasting. International economic outlook
• Abstracts: Learning from the ad or relying on related attitudes: the moderating role ofinvolvement. Mother logit analysis of substitution effects in consumer shopping destination choice
• Abstracts: Managing two fits of strategic human resource management. Development and application of a model of personal control in organizations
• Abstracts: The generalized Stein - Rubinstein covariance formula and its application to estimate real systematic risk

This website is not affiliated with document authors or copyright owners. This page is provided for informational purposes only. Unintentional errors are possible.