| fit.seMPP {QRMlib} | R Documentation |
fits marked self-exciting process to a point process object of class MPP
fit.seMPP(PP, markdens = "GPD", model = "Hawkes", mark.influence = TRUE, predictable = FALSE, std.errs = FALSE)
PP |
a point process object of class MPP |
markdens |
name of density of mark distribution; currently must be "GPD" |
model |
name of self-exciting model: Hawkes or ETAS |
mark.influence |
whether marks of marked point process may influence the self-excitement |
predictable |
whether previous events may influence the scaling of mark distribution |
std.errs |
whether standard errors should be computed VALUE |
see pages 307-309 of QRM
a fitted self-exciting process object of class sePP
fit.sePP,
plot.sePP,
stationary.sePP
data(sp500);
sp500.nreturns <- -mk.returns(sp500);
window <- (seriesPositions(sp500.nreturns) >
timeDate("12/31/1995",format = "%m/%d/%Y"));
sp500.nreturns <- sp500.nreturns[window];
tmp <- extremalPP(sp500.nreturns,ne=100);
mod3a <- fit.seMPP(tmp,mark.influence=FALSE,std.errs=TRUE);
## Not run:
mod3b <- fit.seMPP(tmp,mark.influence=TRUE,std.errs=TRUE);
mod3c <- fit.seMPP(tmp,model="ETAS",mark.influence=FALSE,std.errs=TRUE);
mod3d <- fit.seMPP(tmp,model="ETAS",mark.influence=TRUE,std.errs=TRUE);
mod4a <- fit.seMPP(tmp,mark.influence=FALSE,predictable=TRUE,
std.errs=TRUE);
mod4b <- fit.seMPP(tmp,mark.influence=TRUE,predictable=TRUE,
std.errs=TRUE);
mod4c <- fit.seMPP(tmp,model="ETAS",mark.influence=FALSE,
predictable=TRUE,std.errs=TRUE);
mod4d <- fit.seMPP(tmp,model="ETAS",mark.influence=TRUE,
predictable=TRUE,std.errs=TRUE);
## End(Not run)