predict.systemfit: Predictions from System Estimation

Description

Returns the predicted values, their standard errors and the confidence limits of prediction.

Usage

# S3 method for systemfit
predict( object, newdata = NULL,
                             se.fit = FALSE, se.pred = FALSE,
                             interval = "none", level=0.95,
                             useDfSys = NULL, ... )
# S3 method for systemfit.equation
predict( object, newdata = NULL,
                             se.fit = FALSE, se.pred = FALSE,
                             interval = "none", level=0.95,
                             useDfSys = NULL, ... )

Value

predict.systemfit returns a dataframe that contains for each equation the predicted values ("<eqnLable>.pred") and if requested the standard errors of the fitted values ("<eqnLable>.se.fit"), the standard errors of the prediction ("<eqnLable>.se.pred"), and the lower ("<eqnLable>.lwr") and upper ("<eqnLable>.upr") limits of the confidence or prediction interval(s).

predict.systemfit.equation returns a dataframe that contains the predicted values ("fit") and if requested the standard errors of the fitted values ("se.fit"), the standard errors of the prediction ("se.pred"), and the lower ("lwr") and upper ("upr") limits of the confidence or prediction interval(s).

Arguments

object: an object of class systemfit or systemfit.equation.
newdata: An optional data frame in which to look for variables with which to predict. If it is NULL, the fitted values are returned.
se.fit: return the standard error of the fitted values?
se.pred: return the standard error of prediction?
interval: Type of interval calculation ("none", "confidence" or "prediction")
level: Tolerance/confidence level.
useDfSys: logical. Use the degrees of freedom of the whole system (in place of the degrees of freedom of the single equation) to calculate the confidence or prediction intervals. If it not specified (NULL), it is set to TRUE if restrictions on the coefficients are imposed and FALSE otherwise.
...: additional optional arguments.

Author

Arne Henningsen arne.henningsen@googlemail.com

Details

The variance of the fitted values (used to calculate the standard errors of the fitted values and the "confidence interval") is calculated by $V a r [E [y^{0}] - {\hat{y}}^{0}] = x^{0} V a r [b] {x^{0}}^{'}$
The variances of the predicted values (used to calculate the standard errors of the predicted values and the "prediction intervals") is calculated by $V a r [y^{0} - {\hat{y}}^{0}] = {\hat{σ}}^{2} + x^{0} V a r [b] {x^{0}}^{'}$

References

Greene, W. H. (2003) Econometric Analysis, Fifth Edition, Macmillan.

Gujarati, D. N. (1995) Basic Econometrics, Third Edition, McGraw-Hill.

Kmenta, J. (1997) Elements of Econometrics, Second Edition, University of Michigan Publishing.

Examples

Run this code

data( "Kmenta" )
eqDemand <- consump ~ price + income
eqSupply <- consump ~ price + farmPrice + trend
system <- list( demand = eqDemand, supply = eqSupply )

## OLS estimation
fitols <- systemfit( system, data=Kmenta )

## predicted values and limits
predict( fitols )

## predicted values of the first equation
predict( fitols$eq[[1]] )

## predicted values of the second equation
predict( fitols$eq[[2]] )

Run the code above in your browser using DataLab

The Learning Leader's Guide to AI Literacy