This function performs an heuristic model search, to select which type of sender and receiver effects should be included in the model. The sender/receiver effects can be either absent, constant or variable.
heuSearch( Y, D = 2, nrep = 1000)A three-dimensional array or list of \((n\times n)\) adjacency matrices composing the multidimensional network. A list will be converted to an array. If an array, the dimension of Y must be (n,n,K), where n is the number of nodes and K the number of networks. Structural missing values (NA) are allowed, but no imputation is performed; see details.
The dimension of the latent space, with D > 0. The default value is D = 2.
The number of datasets to be simulated for each model type. By default nrep = 1000.
The function returns a list containing the following objects:
bestModel The selected combination of sender and receiver effects.
modelProbs The estimated probabilities that Y has been generated by the models corresponding to all the sender/receiver effects combinations.
modClass The model-specific accuracies of the classifier. Rows correspond to the predicted model, while columns to the model the simualted data have been generated from.
The function performs an heuristic model search for the type of sender/receiver effects to be included in the model for the multidimensional network Y with \(n\) nodes and \(K\) networks.
The function simulates nrep different multidimensional networks of dimensions \((n, K)\) from the nine models arising from different combinations of sender/receiver effects: "NN", "CN", "NC", "CC", "VN", "NV", "VC", "CV", "VV". The first letter indicates the sender effect type, while the second letter denotes the receiver effect type. The effects can be constrained across the networks, with "N" indicating no effects, "C" constant effects and "V" variable effects. The simulated multidimensional networks are then used to train a classifier, which is employed to assign the observed multiplex Y to the most plausible node-specific effect model.
D'Angelo, S. and Murphy, T. B. and Alf<U+00F2>, M. (2018). Latent space modeling of multidimensional networks with application to the exchange of votes in the Eurovision Song Contest. arXiv.
D'Angelo, S. and Alf<U+00F2>, M. and Murphy, T. B. (2018). Node-specific effects in latent space modelling of multidimensional networks. arXiv.
# NOT RUN {
data(vickers)
#heuS <- heuSearch( vickers, D = 2, nrep = 10) # nrep is set to 10 as an example.
# }
Run the code above in your browser using DataLab