« Sales forecasting and direct mail optimization with Revolution R Enterprise and Alteryx | Main | Atlantic Feature on AI and R »

February 18, 2014


Feed You can follow this conversation by subscribing to the comment feed for this post.

compared with the Figure 3 in Cannarella's paper, the fit is not good enough, any solutions to find the best fit?

This google search trend data is standardized, rather than raw counts. What impact would you say this has on the analysis and interpretation here?

The SIR model has three parameters that you can play with - the initial number of infected, beta(infection rate) and gamma(recovery rate). I probably started with a low initial infected number. You can possibly try with a different start and see how it affects the fit. As per Cannerella's paper, IrSIR model offerred a better fit than SIR model. Note that the figure 3 in the paper also has a different aspect ratio and may look like a better fit.

The comments to this entry are closed.

Search Revolutions Blog

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid
Get this blog via email with Blogtrottr