How to keep track of cumulative number of simulated events?

[TDmetric]

"Look at a stone cutter hammering away at his rock, perhaps a hundred times without as much as a crack showing in it. Yet at the hundred-and-first blow it will split in two, and I know it was not the last blow that did it, but all that had gone before."

I saw this quote a couple weeks ago and started thinking about how events like this could be modeled. I think what I'm trying to demo below could be called a joint Compound Poisson Process/Time to event (TTE) model? My issue is, I can't figure out how to keep track of the cumulative number of simulated events (strikes to the rock) to use as a covariate in the TTE model (stone breaking). Any thoughts from the community? Thank you in advance. TD

cptte <- '
$PLUGIN Rcpp

$PARAM @Annotated
// counting process
TVWT : 1/30 : typical wait time between events
TVLAM : 1 : lambda Poisson process

// time to event
TVLAM2 : 0.01 : lamba for exponential time-to-event model

$CMT @Annotated
HAZ : hazard

$ODE
dxdt_HAZ = TVLAM2; // I want the cumulative sum of MAG to be a covariate on the hazard

$TABLE
// counting process
double ranexp = R::rexp(TVWT);
double cdf = R::pexp(ranexp, TVWT, 1, 0);
double ranun = R::runif(0, 1);
double event = ranun < cdf ? 1 : 0;
double MAG = R::rpois(TVLAM*event); // how can I keep track of the total events during the simulation?

// TTE
double CHAZ = HAZ;
double SUR = exp(-CHAZ);
double HAZNOW = TVLAM2;

$CAPTURE ranexp cdf ranun event MAG TVWT TVLAM TVLAM2
'

[kylebaron]

Hi Tyler -

If you just want to keep track of cumulative MAG, you can try this:

• The code in $MAIN both initializes the variable and resets it to zero with each individual
• The code in $TABLE accumulates MAG

Code isn't tested but let me know if this works for you.

$MAIN
if(NEWIND <= 1) double CUMAG = 0;

$TABLE
double MAG = R::rpois(TVLAM*event);
CUMAG = CUMAG + MAG;

[TDmetric]

So simple! It seems to work. Thanks so much. TD