Neural network models for time series forecasts

Article Abstract:

Traditional statistical models and neural networks are compared in terms of their accuracy in time series forecasting. An experiment is conducted in which forecasts generated by neural networks are compared with forecasts produced using six statistical time series methods in the Makridakis competition on level ground. The performance of neural network forecasts is evaluated vis-a-vis the performance of forecasts obtained using traditional statistical time-series methods, such as Box-Jenkins and exponential smoothing, and a judgment-based method as described by Lawrence et al. (1985). Findings show that neural networks generate more accurate forecasts across monthly and quarterly time series. However, the neural network model and traditional models show comparable performance when forecasting annual data.

Computer networks, Neural networks

---

An Urban Allocation Model Combining Time Series and Analytic Hierachical Methods

Article Abstract:

Modeling and forecasting short-term household population growth and location is important to many organizations. Household population is forecast for the Portland SMSA. A model is developed that uses both time series and cross-sectional data. Forecasts are allocated to planning divisions within the Standard Metropolitan Statistical Area (SMSA) by univariate stochastic models.

Growth (Physiology), Modeling, Data modeling software, Growth, Stochastic Model, Urban Planning

---

An Urban Allocation Model Combining Time Series and Analytic Hierarchical Methods

Article Abstract:

A method is developed for forecasting short-term household population. The urban population is considered. The allocation model combines Box-Jenkins methods with the analytical hierarchy process. The model improves the quality of information, analysis and accuracy of forecasts. Tables are included. An appendix gives symbols.

Economic forecasting

Similar abstracts:

• Abstracts: Accuracy analysis of consensus forecasts. The next generation: hopefully better forecasters. Daily demand forecasting at Columbia Gas
• Abstracts: Adaptive change in corporate control practices. Evaluating Commercial Credit Decisions by Analytic Hierarchy Process
• Abstracts: PNG gets serious about corruption. Turmoil puts mining agreements in doubt
• Abstracts: Approximations for a lost-Sales Production-Inventory Control Model with Service Level Constraints. Discrete Approximations of Probability Distributions
• Abstracts: A Case for Cases. Measuring Skill Gains and Attitudes of Adult Writers in Short Courses

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