Added the social distancing measures notebook #26
Conversation
|
Check out this pull request on You'll be able to see Jupyter notebook diff and discuss changes. Powered by ReviewNB. |
| @@ -0,0 +1,386 @@ | |||
| { | |||
There was a problem hiding this comment.
There was a problem hiding this comment.
I have added a new commit wrt this. Have a look.
| @@ -0,0 +1,386 @@ | |||
| { | |||
There was a problem hiding this comment.
Do you think it'd be better to model this in two splits? One relaxed lockdown from April 20 and completely relaxed from May 2?
How do we model these two scenarios? What is Day 0 and Day 25 in real dates?
Reply via ReviewNB
There was a problem hiding this comment.
Q1: Sure, I can incorporate this by relaxing the scale parameter of my graph defined at the starting of the notebook. I can increase the scale from 10 to 20 for a relaxes lock-down type scenario
Q2: The data is for April 18th at the moment. In the new commit, day 0 is April 18th, a lock-down of 15 days till May 3rd and then trend for 85 more days
There was a problem hiding this comment.
How can we rely that these params work for us? Can we use Lombardo or Wuhan for reference and get corresponding Day 0 curves and see how they compare against reality?
| @@ -0,0 +1,386 @@ | |||
| { | |||
There was a problem hiding this comment.
What is the thought process behind using mean of all the simulations? Am I interpreting the code+plot correctly?
Reply via ReviewNB
There was a problem hiding this comment.
The time steps for SEIRSPLUS modelling is less than 1. If we look at model.tseries, we will be able to see the exact time period for the model is predicting infections. Let's say, on day 14, there are multiple time periods for which we have S, E, D_E, D_I, I, R, F with sufficient std dev. Therefore, I take mean to get a stable value for that day.
| @@ -0,0 +1,386 @@ | |||
| { | |||
There was a problem hiding this comment.
There was a problem hiding this comment.
I had used the default graph provided in the SEIRSPLUS library. I have removed this graph altogether in the new commit.
No description provided.