Error incurred plotting the rejection file in R

[federica mattucci]

Hi all,

I modified the reg_step script considering 22 parameters and 30 statistics and the plot_line script works correctly!

i'm trying to plot the rejection script but i incurred this error

> abc.reg.tev1 <- makepd4(target,abc.rej[,7],abc.rej[,23:46],tol=1,rej=F,transf="none",bb=c(mintev1,maxtev1))

Errore in fit1$coeff %*% c(1, target.s) : gli argomenti non sono compatibili

> plot(locfit(~abc.reg.tev1$x,weight=abc.reg.tev1$wt,xlim=c(mintev1,maxtev1)),main="prior (black) and posterior (blue) distributions",xlab="tev1",col="blue")

Errore in eval(expr, envir, enclos) : oggetto "abc.reg.tev1" non trovato

> plot(locfit(~priors[,7],xlim=c(mintev1,maxtev1)),col="black",add=T)

Errore in plot.xy(xy.coords(x, y), type = type, ...) : plot.new non è ancora stato chiamato

Does anyone know how fix this problem?

Thanks for your availability

With best regards

Federica

[Joao Sollari Lopes]

Hi Federica,

The script you mention in your email is the one that performs the regression-step, not the rejection-step.
In any case, the error seems to be related to the fit, are you providing the correct positions of abc.rej? Also, you should try to look at how the simulations fit the target data (check this topic for some help on that).
Best,
Joao