transitions occurring where it's not possible!

[amcgail]

I'm using your algorithms to validate my own.

We agree for the most part on the sampleData you include, but when I use my own dataset, your algorithm has new infections occurring during the nighttime (when no one is interacting)!

I've attached the file I was using: High-School-transform_data_2013.txt

This is the command: ./poo.exe High-School-transform_data_2013.csv 20 0.02 0.00028492214857 18178 500 1
Where poo.exe is "./SIR-Poisson-homogeneous.cpp" compiled.

Let me know if you can make sense of this!

Here's a graph of my simulated disease spread (sum of avg(I_t) and avg(R_t):

And a graph of yours:

[amcgail]

Update: I've figured it out, I believe. You have the first infection starting at a random time in this range.
When I set

// Random starting time of infection:
t_infectionStart=0;

I get a result more agreeable with my own: