R code for the paper - Hermanussen, M., Dammhahn, M., Scheffler C. & Groth. D. Winner-loser effects improve social network efficiency between competitors with equal resource holding power. Sci Rep 13, 14439 (2023). https://doi.org/10.1038/s41598-023-41225-y - or in nature.com https://www.nature.com/articles/s41598-023-41225-y

Installation (version 0.2.1, version forthe paper figures):

> install.packages("https://github.com/mittelmark/hanna/releases/download/0.2.1/hanna_0.2.1.tar.gz")
    repos=NULL)

Alternatively you can install the latest version directly from Github using the remotes library:

> install.libarary(remotes)
> library(remotes)
> remotes::install_github("https://github.com/mittelmark/hanna")

Thereafter you can check the installation like this:

> library(hanna)
> citation("hanna")

Which should display something like this:

> citation("hanna")
To cite package 'hanna' in publications use:

  Detlef Groth, University of Potsdam (2023). hanna:
  Winner-looser effects for social network efficiency simulated using
  Monte Carlo simulations. R package version 0.1.

Beside of the R functions and the manual pages for them are three vignettes which you can browser by using the vignette commands:

> vignette(package="hanna")
Vignettes in package 'hanna':

simul-tutorial          Example simulations for the hanna package using
                        the simul module (source, html)
figures                 Hermanussen et. al. (2023) - Paper figures
                        (source, html)
hgraph-tutorial         hgraph introduction (source, html)

> ### Let's load one vignette from this package
> vignette("figures",package="hanna")

If the installation was successful, you can run a single season giving the agents as many tokens as there are agents in the game using the default null model like this:

> library(hanna)
> res=simul$season(LETTERS[1:6])
> res$token
### $token
### A B C D E F 
### 7 8 4 5 7 5 
> res$M
### $M
###    A  B C  D  E F
### A  0 -1 1  1  0 0
### B  1  0 0  1  0 0
### C -1  0 0 -1  0 0
### D -1 -1 1  0 -1 1
### E  0  0 0  1  0 0
### F  0  0 0 -1  0 0

The token does represent the overall win and loosing points, so for instance A has won 2 times and lost 1 time so it has 6 2-1 token remaining, it is 7. In the null model win/loose changes are independent from the number of tokens. To use other models consult the help page from ?simul_season. More details are as well shown in the package vignette which you can read usually by writing vignette('simul-tutorial') in your R console.

If you have such a season with the embedded result matrix you can visualize this by using the function simul$graph and plotting the resulting adjacency matrix like this:

> A = simul$graph(res$M,model='win')
> par(mfrow=c(1,2),mai=rep(0.4,2))  
> hgraph$plot(A)  
> cols=hgraph$colors(A)
> hgraph$plot(A,vertex.color=cols)

Agents which are winning all games are shown here in red, agents loosing all games would be shown in blue. For more examples on how to do simulations using the other models look at the manual pages and at the package vignettes.

Author: Detlef Groth, University of Potsdam, Germany

License: MIT license, see file LICENSE.

Please use the Issues link on top of this README on Github.