b2ML: MCMC estimates for the (multilevel) b2 model

In dyads: Dyadic Network Analysis

MCMC estimates for the (multilevel) b2 model

Description

Estimates the (multilevel) b2 model parameters, which is the bidirectional counterpart of the multilevel p2 model as described in Zijlstra, Van Duijn and Snijders (2006) <doi: 10.1027/1614-2241.2.1.42>.

Usage

b2ML(nets, actor = NULL, density = NULL, adapt = NULL, burnin = NULL, center = NULL, 
separate= NULL, densVar = NULL, seed = NULL) 

Arguments

nets	List with n dichotomous symmetric dependent networks.
actor	Optional matrix with a stacked actor covariate, corresponding to the n networks. Multiple actor covariates can be added as a formula object, see example below
density	Optional matrix with symmetric a stacked density covariate, with dimensions similar to the n dependent networks. Multiple density covariates can be added as a formula object, see example below
adapt	Optional number of adaptive sequenses (default is 100).
burnin	Optional specification of number of burn-in iterations (default is 5000).
center	Optional argument for centering predictors (default is TRUE).
separate	Optional argument for estimating separate coefficients for the n dependent networks (default is FALSE).
densVar	Optional argument for estimating densty variance at the network level (default is TRUE).
seed	Optonal specification of random seed (delfault is 1).

Value

Returns a fitted model of class b2ML, to be opened with the function summary().

Author(s)

Bonne J.H. Zijlstra [e-mail]

References

Zijlstra, B. J., Van Duijn, M. A., & Snijders, T. A. (2006). The Multilevel p2 Model A random effects model for the analysis of multiple social networks. Methodology: European Journal of Research Methods for the Behavioral and Social Sciences, 2(1), 42.

Examples

# create two very small networks with covariates for illustrative purposes
Y1 <- matrix(c( 0,1,1,1,1,1,1,1,1,0,
                1,0,1,0,1,1,1,1,1,1,
                1,1,0,0,1,1,1,1,0,1,
                1,0,0,0,1,0,0,1,0,0,
                1,1,1,1,0,1,1,0,1,1,
                1,1,1,0,1,0,1,0,1,1,
                1,1,1,0,1,1,0,1,1,1,
                1,1,1,1,0,0,1,0,0,1,
                1,1,0,0,1,1,1,0,0,1,
                0,1,1,0,1,1,1,1,1,0), ncol=10) 
Y2 <- matrix(c( 0,0,1,0,1,1,0,1,0,0,
                0,0,0,0,0,0,0,1,1,0,
                1,0,0,1,0,1,0,1,0,0,
                0,0,1,0,0,0,1,1,0,0,
                1,0,0,0,0,0,1,1,0,0,
                1,0,1,0,0,0,1,1,0,0,
                0,0,0,1,1,1,0,1,0,0,
                1,1,1,1,1,1,1,0,0,1,
                0,1,0,0,0,0,0,0,0,0,
                0,0,0,0,0,0,0,1,0,0), ncol=10) 
Y <- list(Y1, Y2)                
Aa1 <- c(1,0,1,0,1,1,0,1,0,1)
Aa2 <- c(1,0,0,1,0,0,1,1,0,1)
Aa <- list(Aa1, Aa2)
Aat <- do.call(plyr::rbind.fill.matrix, Aa)
Ab1 <- c(0,0,0,0,0,0,0,0,0,0)
Ab2 <- c(1,1,1,1,1,1,1,1,1,1)
Ab <- list(Ab1, Ab2)
Abt <- do.call(plyr::rbind.fill.matrix, Ab)
Da1 <- abs(matrix(rep(Aa1,10), byrow = FALSE, ncol= 10) -
            matrix(rep(Aa1,10), byrow = TRUE, ncol= 10))
Da2 <- abs(matrix(rep(Aa2,10), byrow = FALSE, ncol= 10) -
            matrix(rep(Aa2,10), byrow = TRUE, ncol= 10))
Da <- list(Da1, Da2)
Dat <- do.call(plyr::rbind.fill.matrix, Da)

# estimate b2 model for two networks
M1 <- b2ML(Y,actor= ~ Aat + Abt, density = ~ Dat, adapt = 10, burnin = 100, densVar = FALSE)
summary(M1)
# Notice: burn-in, and number of adaptive sequenses are 
# much smaller than recommended to keep computation time low.
# recommended code: 
## Not run: 
M1 <- b2ML(Y,actor= ~ Aat + Abt, density = ~ Dat, densVar = FALSE)
summary(M1)

## End(Not run)

# estimate b2 model for a single network
M2 <- b2ML(list(Y[[1]]),actor= ~ Aat[1:10,], density = ~ Dat[1:10,], adapt = 10, burnin = 100, 
densVar = FALSE)
summary(M2)
# Notice: burn-in, and number of adaptive sequenses are 
# much smaller than recommended to keep computation time low.
# recommended code: 
## Not run: 
M2 <- b2ML(list(Y[[1]]),actor= ~ Aat[1:10,], density = ~ Dat[1:10,], densVar = FALSE)
summary(M2)

## End(Not run)

dyads documentation built on Aug. 17, 2022, 9:06 a.m.