| Title: | Simulate Spatial Bernoulli Networks |
|---|---|
| Description: | Social network analysis is becoming commonplace in many social science disciplines, but access to useful network data, especially among marginalized populations, still remains a formidable challenge. This package mitigates that problem by providing tools to simulate spatial Bernoulli networks as proposed in Carter T. Butts (2002, ISBN:978-0-493-72676-2), "Spatial models of large-scale interpersonal networks." Using this package, network analysts can simulate a spatial point process or sequence with a given number of nodes inside a geographical boundary and estimate the probability of a tie formation between all node pairs. When simulating a network, an analyst can choose between five spatial interaction functions. The package also enables quick comparison of summary statistics for simulated networks and provides simple to use plotting methods for its classes that return plots which can be further refined with the 'ggplot2' package. |
| Authors: | Darren Colby [aut, cre] |
| Maintainer: | Darren Colby <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.2.1 |
| Built: | 2025-01-20 04:12:02 UTC |
| Source: | https://github.com/dscolby/spacejamr |
Creates a new spacejamr object that for further analysis
as.spacejamr(path, guess_crs = TRUE)as.spacejamr(path, guess_crs = TRUE)
path |
the path to a shapefile as a string. |
guess_crs |
whether to try to guess the coordinate reference system of the supplied shapefile |
The returned spacejamr object will contain a window object containing a geographical boundary and its coordinate reference system. Since any simulated point process will be simulated in two dimensions, the coordinate reference system of the supplied shapefile should be a projected coordinate reference system or 'guess_crs' should be set to TRUE. In that case, the coordinate reference system will be set to the most appropriate coordinate reference system for the shapefile. Note that guessing the coordinate reference system will take longer than if one is already set.
a spacejamr object containing two items. window: object of class 'owin' that stores geographical boundaries. crs: integer value referring to the coordinate reference system of the geographical boundaries.
Darren Colby
Email: [email protected]
ri <- as.spacejamr(system.file("shape/ri.shp", package = "spacejamr"))ri <- as.spacejamr(system.file("shape/ri.shp", package = "spacejamr"))
Generates summary statistics for two simulated networks of class NetSim.
compare_networks(net1, net2)compare_networks(net1, net2)
net1 |
a NetSim object |
net2 |
a NetSim object |
This function enables a quick comparison of two simulated networks of class NetSim. This allows the user to see how different a network simulated from a Halton sequence is from network simulated from a Poisson point process with the same number of points in the same geographical window. Similarly, it enables comparison of networks generated from different spatial interaction functions. The density, mean degree, mean closeness, mean betweenness, and the size of the largest connected component are computed for each network.
A dataframe with two rows and five columns. The first row displays statistics from the first network and the second row displays statistics from the second network. The values in each column are as follows. Density: the density of the network. Mean Degree: the average degree of all nodes in the network. Mean Closeness: the average closeness centrality of all nodes in each network. Mean Betweenness: the average betweenness centrality for the nodes in a network. Largest Component Size: The number of nodes in the largest connected component of a network.
Darren Colby
Email: [email protected]
# load data data("RI") # Simulate point processes points1 <- PointSim(10, RI) points2 <- PointSim(20, RI) # Create two networks net1 <- NetSim(points1) net2 <- NetSim(points2) compare_networks(net1, net2)# load data data("RI") # Simulate point processes points1 <- PointSim(10, RI) points2 <- PointSim(20, RI) # Create two networks net1 <- NetSim(points1) net2 <- NetSim(points2) compare_networks(net1, net2)
Simulates a spatial Bernoulli network from a NetSim object using one of six probability law distributions.
NetSim(point_sim, sif, base_prob, scale, threshold, power)NetSim(point_sim, sif, base_prob, scale, threshold, power)
point_sim |
a PointSim object |
sif |
the spatial interaction function to use. Use attenuated to use an attenuated power law; arctan to use an arctangent probability law; decay to use an exponential decay law; or logistic to use a logistic probability law. Default is a standard power law function. |
base_prob |
the theoretical probability that two nodes (points) with distance 0 share a tie. Default is 0.9. |
scale |
a coefficient to multiply the distance by. Default is 1. |
threshold |
if two node exceed this probability, they will be coded as having a tie. Default is 0.5. |
power |
the exponent at which tie probability decays. Default is -2.8. |
The algorithm proceeds in three steps. First, it calculates the distance between simulated points from a PointSim object. Then it calculates the distance between all pairs of points. Finally, it uses a spatial interaction function to calculate that any two point share a tie. If the threshold is exceeded, a tie is created.
A network object of class 'NetSim' and 'igraph' that can be manipulated using the igraph' package.
Darren Colby
Email: [email protected]
Butts, Carter T. Spatial models of large-scale interpersonal networks. Dissertation. (2002).
# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) power_law <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4)# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) power_law <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4)
This can take either a PowerLawNetwork or APLNetwork object as input, both of which are chidren of the NetSim class.
## S3 method for class 'NetSim' plot( x, y, ..., layout = "stress", title = "Network Simulation", node_color = "red", edge_color = "blue" )## S3 method for class 'NetSim' plot( x, y, ..., layout = "stress", title = "Network Simulation", node_color = "red", edge_color = "blue" )
x |
a NetSim graph |
y |
ignored. |
... |
ignored. |
layout |
a layout to display the graph. Layout must be a valid string. from the ggraph package. Default is "stress". |
title |
an optional title. |
node_color |
a color for the nodes. Default is blue. |
edge_color |
a color for the edges. Default is red. |
This method returns a ggraph object, which can be further refined using standard ggraph and ggplot facilities.
A plot of classes 'ggraph' 'gg' and 'ggplot'
Darren Colby
Email:[email protected]
# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) spl_points <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4) plot(spl_points)# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) spl_points <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4) plot(spl_points)
Plots the results of points simulated in a PointProcess or HaltonSeq class, whcih obht inherit methods from the PointSim class.
## S3 method for class 'PointSim' plot(x, y, ..., title = "Simulated Points", color = "red")## S3 method for class 'PointSim' plot(x, y, ..., title = "Simulated Points", color = "red")
x |
an object of class PointSim or one of its child classes |
y |
ignored |
... |
ignored |
title |
an optional title. Default is "Simulated Points". |
color |
an optional color for the simulated points. Default is red. |
The returned plot can be refined with standard ggplot2 functions
A plot of classes 'gg' and 'ggplot'
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) plot(ri_points)# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) plot(ri_points)
Plot method for the spacejamr class
## S3 method for class 'spacejamr' plot(x, y, ..., title = "Spatial Window", fill = "blue")## S3 method for class 'spacejamr' plot(x, y, ..., title = "Spatial Window", fill = "blue")
x |
an object of class spacejamr. |
y |
ignored. |
... |
ignored. |
title |
an optional title for the plot. Default is "Spatial Window". |
fill |
an optional fill for the plot. Default is blue. |
The returned plot can be refined with standard ggplot2 functions
A plot of classes 'gg' and 'ggplot'
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") plot(RI)# Load spacejamr object data("RI") plot(RI)
Creates a new Poisson Point Process in a spacejamr object
PointSim(points, window, type, seed)PointSim(points, window, type, seed)
points |
the number of points to simulate |
window |
a spacejamr object to use as the spatial extent |
type |
the type of simulation. poisson_process simulates a spatial Poisson process. halton simulates a Halton sequence where the maximal number of simulated points is the first argument. |
seed |
an optional seed |
An object of class PointSim that contains a geographical window of class 'owin'. Within this window are four objects. n: the number of simulated points. x: the x coordinates of the simulated points. y: the y coordinates of the simulated points. markformat: an empty place holder.
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") # Poisson process ri_points <- PointSim(points = 10, window = RI, seed = 42)# Load spacejamr object data("RI") # Poisson process ri_points <- PointSim(points = 10, window = RI, seed = 42)
Plots a NetSim object and returns a ggraph object
## S3 method for class 'NetSim' print(x, ...)## S3 method for class 'NetSim' print(x, ...)
x |
a NetSim object |
... |
ignored. |
No return value, called for side effects
Darren Colby
Email: [email protected]
# Create spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) spl_points <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4) print(spl_points)# Create spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) spl_points <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4) print(spl_points)
Print method for both the PointProcess and HaltonSeq classes, which inherit methods from the PointSim class.
## S3 method for class 'PointSim' print(x, ...)## S3 method for class 'PointSim' print(x, ...)
x |
a PointSim object or a child object |
... |
ignored. |
No return value, called for side effects
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) print(ri_points)# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) print(ri_points)
Print method for the spacejamr class
## S3 method for class 'spacejamr' print(x, ...)## S3 method for class 'spacejamr' print(x, ...)
x |
a spacejamr object |
... |
ignored. |
Provides a wrapper for the print.owin method in the spatstats.geom package.
No return value, called for side effects
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") print(RI)# Load spacejamr object data("RI") print(RI)
A dataset containing Rhode Island's border and the coordinate reference system in which it is projected.
RIRI
A spacejamr object with an owin window object and an integer denoting the projected coordinate reference system:
geographical boundary of Rhode Island
the EPSG code for the coordinate reference system of the window
...
https://www.rigis.org/datasets/edc::state-boundary-1989/about
The spacejamr package provides three main types of functions: spacejamr, PointSim, and NetSim. They create new objects used in subsequent steps of network simulation, simulate spatial point processes, and create network objects. There are also plot, print, and summary methods for all objects and a compare_networks function to quickly compare two simulated networks.
APLNetwork
compare_networks
HaltonSeq
RI
plot.NetSim
plot.PointSim
plot.spacejamr
PointProcess
PowerLawNetwork
print.NetSim
print.PointSim
print.spacejamr
spacejamr
summary.NetSim
summary.spacejamr
summary.PointSim
Darren Colby
Email: [email protected]
ORCID: 0000-0001-8468-2755
Prints a summary of a NetSim object
## S3 method for class 'NetSim' summary(object, ...)## S3 method for class 'NetSim' summary(object, ...)
object |
a NetSim object |
... |
ignored. |
No return value, called for side effects
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) spl_points <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4) summary(spl_points)# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) spl_points <- NetSim(ri_points, base_prob = 0.92, scale = 1, threshold = 0.5, power = -2.4) summary(spl_points)
Prints a summary of information from either a PointProcess or HaltonSeq object, whcih are both child classes of the PointSim class.
## S3 method for class 'PointSim' summary(object, ...)## S3 method for class 'PointSim' summary(object, ...)
object |
a PointSim object |
... |
ignored |
No return value, called for side effects
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) summary(ri_points)# Load spacejamr object data("RI") ri_points <- PointSim(points = 10, window = RI, seed = 42) summary(ri_points)
Summary method for the spacejamr class
## S3 method for class 'spacejamr' summary(object, ...)## S3 method for class 'spacejamr' summary(object, ...)
object |
a spacejamr object |
... |
ignored. |
Provides a wrapper for the summary.owin method in the spatstats.geom package.
No return value, called for side effects
Darren Colby
Email: [email protected]
# Load spacejamr object data("RI") summary(RI)# Load spacejamr object data("RI") summary(RI)