plot.PAFit_result: Plotting the estimated attachment function and node fitness...

In PAFit: Generative Mechanism Estimation in Temporal Complex Networks

Plotting the estimated attachment function and node fitness of a PAFit_result object

Description

This function plots the estimated attachment function A_k and node fitness eta_i, together with additional information such as their confidence intervals or the estimated attachment exponent (\alpha when assuming A_k = k^\alpha) of a PAFit_result object. This object is stored in the field $estimate_result of a Full_PAFit_result object, which in turn is the returning value of only_A_estimate, only_F_estimate or joint_estimate.

Usage

  ## S3 method for class 'PAFit_result'
plot(x,
    net_stat       = NULL    ,
    true_f         = NULL    , plot             = "A"              , plot_bin   = TRUE ,
    line           = FALSE   , confidence       = TRUE             , high_deg_A = 1    ,
    high_deg_f     = 5       ,
    shade_point    = 0.5     , col_point        = "grey25"         , pch        = 16   ,
    shade_interval = 0.5     , col_interval     = "lightsteelblue" , label_x    = NULL , 
    label_y        = NULL    ,
    max_A          = NULL    , min_A            = NULL             , f_min      = NULL , 
    f_max          = NULL    , plot_true_degree = FALSE , 
    ...)

Arguments

x	An object of class PAFit_result.
net_stat	An object of class PAFit_data, containing the summerized statistics.
true_f	Vector. Optional parameter for the true value of node fitnesses (only available in simulated datasets). If this parameter is specified and plot == "true_f", a plot of estimated \eta versus true \eta is produced (after a suitable rescaling of the estimated f).
plot	String. Indicates which plot is produced. If "A" then PA function is plotted. If "f" then the histogram of estimated fitness is plotted. If "true_f" then estimated fitness and true fitness are plotted together (require supplement of true fitness). Default value is "A".
plot_bin	Logical. If TRUE then only the center of each bin is plotted. Default is TRUE.
line	Logical. Indicates whether to plot the line fitted from the log-linear model or not. Default value is TRUE.
confidence	Logical. Indicates whether to plot the confidence intervals of A_k and eta_i or not. If confidence == TRUE, a 2-sigma confidence interval will be plotted at each A_k and eta_i.
high_deg_A	Integer. The estimated PA function is plotted starting from high_deg_A. Default value is 1.
high_deg_f	Integer. If plot == "true_f", only nodes whose number of edges acquired is not less than high_deg_f are plotted. Default value is 5.
col_point	String. The name of the color of the points. Default value is "black".
shade_point	Numeric. Value between 0 and 1. This is the transparency level of the points. Default value is 0.5.
pch	Numeric. The plot symbol. Default value is 16.
shade_interval	Numeric. Value between 0 and 1. This is the transparency level of the confidence intervals. Default value is 0.5.
max_A	Numeric. Specify the maximum of the axis of PA.
min_A	Numeric. Specify the minimum of the axis of PA.
f_min	Numeric. Specify the minimum of the axis of fitness.
f_max	Numeric. Specify the maximum of the axis of fitness.
plot_true_degree	Logical. The degree of each node is plotted or not.
label_x	String. The label of x-axis.
label_y	String. The label of y-axis.
col_interval	String. The name of the color of the confidence intervals. Default value is "lightsteelblue".
...	Other arguments to pass to the underlying plotting function.

Value

Outputs the desired plot.

Author(s)

Thong Pham [e-mail]

Examples

  ## Since the runtime is long, we do not let this example run on CRAN
  ## Not run: 
    library("PAFit")
    set.seed(1)
    # a network from Bianconi-Barabasi model
    net        <- generate_BB(N        = 1000 , m             = 50 , 
                              num_seed = 100  , multiple_node = 100,
                              s        = 10)
    net_stats  <- get_statistics(net)
    result     <- joint_estimate(net, net_stats)
    #plot A
    plot(result$estimate_result , net_stats , plot = "A")
    true_A     <- c(1,result$estimate_result$center_k[-1])
    lines(result$estimate_result$center_k + 1 , true_A , col = "red") # true line
    legend("topleft" , legend = "True function" , col = "red" , lty = 1 , bty = "n")
    #plot true_f
    plot(result, net_stats , net$fitness, plot = "true_f")
  
## End(Not run)

PAFit documentation built on June 22, 2024, 11:06 a.m.