Browse Prior Art Database

Forecasting and Simulating With Econometric Nonlinear Models

IP.com Disclosure Number: IPCOM000076444D
Original Publication Date: 1972-Mar-01
Included in the Prior Art Database: 2005-Feb-24
Document File: 3 page(s) / 21K

Publishing Venue

IBM

Related People

Chignoli, C: AUTHOR [+3]

Abstract

A powerful tool in forecasting and simulating an off-line estimated econometric model is provided by a package including the following features: 1) Processing any kind of nonlinearities in the exogenous variables. 2) Solution of nonlinearities in the endogenous variables with an algorithm which reduces the size of the initial nonlinear system without reducing, under some conditions, its rate of convergence, thus decreasing the over-all processing time. 3) Exchange of the role between endogenous and and exogenous variables. 4) Lag correspondence extended to any number of time intervals. 5) Parametric solution. 6) Matrix reconditioning. 7) Macro-instruction to iterate simulation under different conditions. 8) Verification of the model versus the time series.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 56% of the total text.

Page 1 of 3

Forecasting and Simulating With Econometric Nonlinear Models

A powerful tool in forecasting and simulating an off-line estimated econometric model is provided by a package including the following features:
1) Processing any kind of nonlinearities

in the exogenous variables.
2) Solution of nonlinearities in the endogenous variables with an

algorithm which reduces the size of the initial

nonlinear system without reducing, under

some conditions, its rate of convergence,

thus decreasing the over-all processing time.
3) Exchange of the role between endogenous and

and exogenous variables.
4) Lag correspondence extended to any number of time intervals.
5) Parametric solution.
6) Matrix reconditioning.
7) Macro-instruction to iterate

simulation under different conditions.
8) Verification of the model versus the time series.
9) Display the inverse matrix and the matrix of the multipliers.

Details relating to features 1 and 2 follow.

Feature 1. It is based on a parser which

is able to build polish strings out of

any FORTRAN - like expression describing a

nonlinear relationship among the variables. Let the system be:
(1) f(1) (y(1),..-,y(n) x(1) ... x(n)) = 0 i=l ...,n where y3s are endogenous variables at time t and x3s are lagged endogenous variables or exogenous variables. With nonlinearities in the exogenous variables is meant any functional relation with the only restriction

(Image Omitted)

Feature 2. The nonlinear system (1) of n equations is linearized by introducing...