search

alsnhll / SEIR_COVID19_Dev

(Developers version) SEIR model for COVID-19 infection, including different clinical trajectories of infection
17 stars 16 forks source link

Include seasonality #5

Closed alsnhll closed 4 years ago

mzlevy commented 4 years ago

We don't have great reason to hope for seasonality (in year 1--i think maybe we do expect it in subsequent years) but want to include sliders so we can play out scenarios. To set defaults could use: For northern temperate regions: ε seasonal forcing: uniform .25-.75 θ peak transmissibility: normal peak at 0; sd .1 taken from the bullet points on page five of neher et al. Mike Neher_seasonality.pdf

mzlevy commented 4 years ago

epsilon<-.5 theta<-.1 in model: seasonality<- (1 + epsilon cos(2pi(t-theta))) b1<-b1seasonality b2<-b2seasonality b3<-b3seasonality

alsnhll commented 4 years ago

There is now a script to runSpreadSeason.R that runs the dynamics with seasonal variation in the Beta values (same for all Betas). Note that time is in days so the correct function is: (1 + seas.ampcos(2pi*(t-seas.phase)/365)) Seasonality affects all Beta values equally right now