Subtitle1: The Science of Multivariate Spectral Forecasting.
Subtitle2: For Turning Points and Changes in Trend.
Includes : Time Series Analysis & Forecasting
teaching computer software (FOURCAST).
Detailed description:
FOURCAST is a computer program for forecasting either univariate or mulivariate time series (a collection of observations made sequentially in time). It is designed to address a broad spectrum of uses in science, finance and industry. Popular applications include index values, price, income, power system loading, sales, gross national product, temperature, rainfall and energy. FOURCAST will process data for multivariate analysis, using several time series, regarding them as auto correlated (the association of each variable with itself over time) and cross correlated (the association between different variables).
FOURCAST analyzes & forecasts times series: . Hourly . Daily . Weekly . Monthly . Bi-Monthly . Quarterly . Bi-Annually . Yearly
SIMPLE TO USE AND FLEXIBLE: . Pop-up main and sub menus . Keyboard macro automation . Telecommunications . Easy updating . Prompt instructed interactive input . A user oriented editor . Data transformation functions . Selective printing and plotting
OPTIONAL REPORT TYPES: . Color graphics forecast . Tabular forecast . Analysis of variance . Parameter estimates . Inferential statistics . Analysis of residuals . Comparison of actual and forecast . Frequency and Impulse response plots
FOURCAST employs a technique known as the Moving Window Spectral Method (MWS). This method is particularly effective when the time series contain cyclical components. A moving time window is used to determine how the cycles change over time. The cycles are forecast individually, then recombined. The forecast is dynamic. When the forecast cycles are in alignment, turning points and changes in trend will occur. Also, antithetic forecasting is used to reduce bias.
