mountains: Mountain plots for summarizing forecast densities

In MSBVAR: Markov-Switching, Bayesian, Vector Autoregression Models

Description

"Mountain plots" summarize the bivariate density of 2 variables for two competing forecasts of those variables.

Usage

mountains(fcasts1, fcasts2, varnames, pts, ...)

Arguments

fcasts1	gibbs x 2 set of forecasts from model 1
fcasts2	gibbs x 2 set of forecasts from model 2
varnames	c("name1","name2") object of the variable names
pts	c(pt1,pt2) which are reference points to be plotted.
...	Other graphics parameters.

Details

A "mountain plot" provide a 2 x 2 graph of plots that summarize the bivariate forecasts for two competing forecasts. This function presents four perspectives on the bivariate density or 'hills' for a set of forecasts. Starting from the bottom right plot and working counter-clockwise, the first plot is the bivariate density of the two competing forecasts. The next plot is a contour map that provide the topography of the densities. The third and fourth plots are projections of densities in each variable. The first forecast in the function is presented in black, the second in red. The densities are estimated from the Gibbs Monte Carlo sample of forecasts using the bkde2D bivariate kernel density estimator with an optimal plug-in bandwidth selected using dpill.

Value

None. Produces the mountain plot described above in the current graphics device.

Note

This function requires the bivariate kernel smoother in the package bkde2D

Author(s)

Patrick T. Brandt

References

Brandt, Patrick T. and John R. Freeman. 2006. "Advances in Bayesian Time Series Modeling and the Study of Politics: Theory Testing, Forecasting, and Policy Analysis" Political Analysis 14(1):1-36.

See Also

bkde2D for details of the density estimators

Examples

## Not run: 
data(IsraelPalestineConflict)

# Fit a BVAR model
fit.BVAR <- szbvar(IsraelPalestineConflict, p=6, z=NULL, lambda0=0.6,
                   lambda1=0.1, lambda3=2, lambda4=0.5, lambda5=0,
                   mu5=0, mu6=0, nu=3, qm=4, prior=0,
                   posterior.fit=FALSE)

# Fit a flat prior / MLE model
fit.FREQ <- szbvar(IsraelPalestineConflict, p=6, z=NULL, lambda0=0.6,
                   lambda1=0.1, lambda3=2, lambda4=0.5, lambda5=0,
                   mu5=0, mu6=0, nu=3, qm=4, prior=2,
                   posterior.fit=FALSE)

# Generate unconditional forecasts for both models
forecast.BVAR <- uc.forecast.var(fit.BVAR, nsteps=2,
                                 burnin=100, gibbs=1000)

forecast.FREQ <- uc.forecast.var(fit.FREQ, nsteps=2,
                                 burnin=100, gibbs=1000)

# Plot the densities for the forecasts in period of the forecast horizon

mountains(forecast.BVAR$forecast[,2,1:2],
  forecast.FREQ$forecast[,2,1:2], varnames=c("I2P","P2I"), pts=c(0,0))

## End(Not run)

MSBVAR documentation built on May 30, 2017, 1:23 a.m.