ashokkrish / episim

episim is an R Shiny app for mathematical modelling of infectious diseases
GNU General Public License v3.0
3 stars 3 forks source link

Implement stochastic epidemic models #96

Open ashokkrish opened 1 month ago

ashokkrish commented 1 month ago

@kle6951

Implement Stochastic epidemic models.

For a Stochastic SIR model I am expecting to see a plot on the mainPanel() looking like this

image

Key References (see Google Drive):

Stochastic epidemic models - a survey - Britton - 2010

kle6951 commented 1 month ago

@ashokkrish I have pushed the code. Please let me know any further information and to-do things.

kle6951 commented 1 month ago

@ashokkrish For the binomial model, I would love to have your verification for this:

Screenshot 2024-06-12 at 4 52 24 PM

May I know what the letter T stands for which variable?

Thank you for the help!

ashokkrish commented 1 month ago

@ashokkrish For the binomial model, May I know what the letter T stands for which variable?

@kle6951 T refers to the Number of Timesteps (m)

kle6951 commented 3 weeks ago

Hi @ashokkrish , for the stochastic models, since you requested to hide the vital dynamic option for the stochastic option before, would you clarify whether we should include the vital dynamic option for stochastic models again? If yes, I will proceed to make changes to the code accordingly. At this moment, I only refactored the code and left the solver separated from the app. Once I have your confirmation, I will link it to the app.

One thing that I need your clarification In the code you provided for binomial SI with vital dynamic, there is only one mu value, which doesn't specify whether it is mu birth or mu death. Therefore, I would love to have your clarification.

Thank you for your help!

bryce-carson commented 2 weeks ago

@kle6951, is this issue closable? If you're working on other stochastic models, please add a task list to this issue like below:

kle6951 commented 2 weeks ago

According to Ashok, the implementations for the stochastic option for SEI-type models and the uniform TMA SIR will be future work. I will leave this issue open as Ashok may want to add something to the stochastic models. However, at this point, both binomial solver(SIR and SIRS) and PMA uniform solver(SIR) are completed based on the resources provided by Ashok.