R/ACD_Class_Fit.r

In fACD: Autoregressive Conditional Duration Models in R

# R file for ACD_Fits S4 object definition and implementation

setClass("acdModel",representation(
                    x="matrix"          ,
                    pLag="numeric"      ,
                    qLag="numeric"      ,
                    condDur="matrix"    ,   # filtered probabilities (tŚt)
                    Coeff="list"        ,   # all coefficients
                    Coeff_Std="list"    ,   # Standard errors for coefficients
                    Coeff_pValues="list",
                    LL="numeric"        ,   # final log likelihood
                    paramVec="matrix"   ,   # vector of parameters (has the same values as Coeff)
                    nParameter="numeric",   # number of parameters in the model
                    sizeModel="list"    ,   # a list with the size of the model (number of indep var, etc)
                    distrib="character" ,
                    typeACD='character' ,
                    timeRun='character' ),
                    
                    package='fACD')
                    
print.acdModel<-function(modelOut)
{
S3method(print, acdModel)

cat('\nPrinting ACD Model\n')
cat('\nType of ACD model:',modelOut@typeACD)
cat('\nDistribution Assumption:',modelOut@distrib)
cat('\nDate and Time of Estimation:',modelOut@timeRun)

cat('\nLag Structure:\n')
cat('   q lags=',modelOut@sizeModel$qLag,'\n')
cat('   p lags=',modelOut@sizeModel$pLag)
cat('\nNumber of Parameters:',modelOut@sizeModel$nParameters)
cat('\nFinal log Likelihood:',modelOut@LL)

cat('\n\nFitted Parameters - Coefficient (standard error, p value):')
cat('\nOmega:',sprintf('%4.4f (%4.2f, %4.4f)',modelOut@Coeff$w,modelOut@Coeff_Std$w,modelOut@Coeff_pValues$w) )

for (i in 1:modelOut@qLag)
    cat('\nAlpha',i,':',sprintf('%4.4f (%4.2f, %4.4f)',modelOut@Coeff$q[(qLag-i+1)],modelOut@Coeff_Std$q[(qLag-i+1)],modelOut@Coeff_pValues$q[(qLag-i+1)]) )
    
for (i in 1:modelOut@pLag)
    cat('\nBeta ',i,':',sprintf('%4.4f (%4.2f, %4.4f)',modelOut@Coeff$p[pLag-i+1],modelOut@Coeff_Std$p[pLag-i+1],modelOut@Coeff_pValues$p[pLag-i+1]) )

if ( (modelOut@typeACD=='BC')|(modelOut@typeACD=='EX'))
{   
    for (i in 1:modelOut@qLag)
        cat('\ndelta',i,':',sprintf('%4.4f (%4.2f, %4.4f)',modelOut@Coeff$delta[qLag-i+1],modelOut@Coeff_Std$delta[qLag-i+1],modelOut@Coeff_pValues$delta[qLag-i+1]) )    
}

if (modelOut@distrib=='weibull')
    cat('\ny:',sprintf('%4.4f (%4.2f, %4.4f)',modelOut@Coeff$y,modelOut@Coeff_Std$y,modelOut@Coeff_pValues$y) )    


}


plot.acdModel<-function(modelOut)
{
    plot(modelOut@x,xlab='Event',ylab='Duration',type='l')
    lines(modelOut@condDur,col='red')
    title('Duration and Conditional Duration')  


}