NlsyLinks (version 2.0.6)

CreateSpatialNeighbours: Distances between related family members, formated for spatial analysis.

Description

This helper function formats the LinksPair datasets so it can be used in some types of spatial analyses. The spdep (Spatial Dependence) uses a sparse matrix (actually a data.frame) to represent neigbours.

Usage

CreateSpatialNeighbours(linksPairsDoubleEntered)

Arguments

linksPairsDoubleEntered

A data.frame containing the links, preferably created by a function like

CreatePairLinksDoubleEntered.

Value

An S3 spatial.neighbours object to work with functions in the spdep package.SubjectTag_S1 is renamed `from'.SubjectTag_S2 is renamed `to'.R is renamed `weight'.The attribute region.id specifies each unique SubjectTag.The attribue n specifies the number of unique subjects.

Details

There is one row per unique pair of subjects, respecting order. This has twice as many rows as Links79Pair and Links79PairExpanded (which have one row per unique pair of subjects, irrespective of order).

CreateSpatialNeighbours accepts any paired relationships in a data.frame, as long as it contains the columns SubjectTag_S1, SubjectTag_S2, and R. See Links79Pair for more details about these columns.

References

Bard, D.E., Beasley, W.H., Meredith, K., & Rodgers, J.L. (2012). Biometric Analysis of Complex NLSY Pedigrees: Introducing a Conditional Autoregressive Biometric (CARB) Mixed Model. Behavior Genetics Association 42nd Annual Meeting. [Slides]

Bivand, R., Pebesma, E., & Gomez-Rubio, V. (2013). Applied Spatial Data Analysis with R. New York: Springer. (Especially Chapter 9.)

Banerjee, S., Carlin, B.P., & Gelfand, A.E. (2004). Hierarchical Modeling and Analysis for Spatial Data. Boca Raton: CRC Press.

Lawson, A.B (2013). Bayesian Disease Mapping: Hierarchical Modeling in Spatial Epidemiology, Second Edition. Boca Raton: CRC Press.

The spdep package documentation: spdep: Spatial dependence: weighting schemes, statistics and models.

Examples

dsLinksAll <- Links79Pair
dsLinksGen1Housemates <- dsLinksAll[dsLinksAll$RelationshipPath=="Gen1Housemates", ]
dsLinksGen2Siblings <- dsLinksAll[dsLinksAll$RelationshipPath=="Gen2Siblings", ]

spGen1 <- CreateSpatialNeighbours(dsLinksGen1Housemates)
spGen2 <- CreateSpatialNeighbours(dsLinksGen2Siblings)

head(spGen2)
#Returns: 
#   from  to weight
#3   201 202   0.50
#6   301 302   0.50
#7   301 303   0.50
#9   302 303   0.50
#24  401 403   0.25
#28  801 802   0.50

table(spGen2$weight)
#Returns:
#0.25 0.375   0.5  0.75     1 
#3442   610  6997    12    27