centralityPlot with confidence intervals about qgraph

[diegogaldinof]

Hi, Arianne!

I guess you need to specify the centrality measures in argument statistics.

Try this:
boot3 <- bootnet(Data, 100, default = "pcor", type = "parametric", statistics = c("edge", "strength", "betweenness", "closeness"))

I hope to be contributing with your analysis.

Cheers!

[ariannechr]

thanks!

…

On Wed, Jun 17, 2020 at 3:49 PM Diego Galdino França < ***@***.***> wrote: Hi, Arianne! I guess you need to specify the centrality measures in argument statistics . Try this: boot3 <- bootnet(Data, 100, default = "pcor", type = "parametric", statistics = c("edge", "strength", "betweenness", "closeness")) I hope to be contributing with your analysis. Cheers! — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#40 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AMOLD6CVJVTRG6FEBKP6ZN3RXFB5ZANCNFSM4KY3EFYQ> .

[alexp872]

Hello!

This fix doesn't work for me - I still get the node strength plot (as in Fig 6, referenced above), but I want the 95% quantile regions plotted in centralityPlot

boot1 <- bootnet(results, nCores = 8,
nBoots = 1000, type = 'nonparametric', statistics= c("edge", "strength","closeness","betweenness"))

plot(boot1, statistics = c("strength"),CIstyle = "quantiles")

Any help much appreciated!

[SachaEpskamp]

This is deliberatly not made possible as bootstrapping CIs around centrality metrics is not possible.