Scale list of binary scans into a weighted adjacency matrix Scaling here is dividing by the sum of 1s by the number of time an edge has been observed (whether it was 0 or 1).

Source: R/expDesign_building.blocks.R

Written as a S3 method to be applied to scanList or sLlist (list of scanList) objects.

scale_scans(scan.list, ...)

Arguments

scan.list

a scanList or sLlist object. See objects returned by simunet()
...

additional arguments to be passed. At the moment scale_scans() does not use

At the moment scale_scans() does not use additional argument, arguments passed will be ignored.

Value

a scaled object, or list of such, consisting mainly on a weighted adjacency matrix where each edge weight is equal to the sum of all binary edges divided by the number of times they have been sampled (determined via count_nonNA()). Inherits from weightedAdj and the previous scanList class (theoretical or empirical, inheriting from scanList), and keeps track of the scan.list's list of attributes attrs

See also

simunet(), design_exp(), perform_exp(), sum_scans().

Examples

set.seed(42)
n <- 5L
samp.effort <- 241L

# Adjacency matrix import
## random directed adjacency matrix
Adj <- sample(1:samp.effort,n * n) |>
  matrix(nrow = 5,dimnames = list(letters[1:n],letters[1:n]))
diag(Adj) <- 0L
Adj

#>     a   b   c   d   e
#> a   0 228  47 165 111
#> b 229   0  24 110 131
#> c  65 122   0  20  41
#> d 153 241 100   0 227
#> e  74 128  89 114   0

# social network simulations
## theoretical scans
sL <- simunet(Adj = Adj,samp.effort = samp.effort,mode = "directed",n.scans = 120L)
sL

#> 
#> scan: 1           
#> a . 1 . 1 .
#> b 1 . . . .
#> c . . . . .
#> d . 1 1 . 1
#> e . . 1 . .
#> 
#> scan: 2           
#> a . 1 . . .
#> b 1 . . 1 .
#> c 1 1 . . .
#> d . 1 . . 1
#> e . . 1 1 .
#> 
#> ... ( 117  more scans)
#> 
#> scan: 120           
#> a . 1 . 1 1
#> b 1 . . 1 .
#> c . 1 . . .
#> d 1 1 . . 1
#> e . 1 1 . .
#> 
#> 
#> Hidden attributes:
#> scanList.type - raw.scanList - Adj - samp.effort - n.scans - mode
#> Adj.subfun - edge.Prob
# scale_scans() can scale `weightedAdj` objects...
sL |> sum_scans() |> scale_scans()

#> 
#> Scaled weighted adjacency matrix                               
#> a .     0.958 0.300 0.525 0.400
#> b 0.933 .     0.092 0.400 0.533
#> c 0.300 0.500 .     0.117 0.142
#> d 0.592 1.000 0.375 .     0.958
#> e 0.242 0.500 0.392 0.450 .    
#> 
#> 
#> Hidden attributes:
#> scanList.type - raw.scanList - Adj - samp.effort - n.scans - mode
#> Adj.subfun - edge.Prob - summed.scanList - sampled


# ... or `scanList` object directly
## group-scan sampling
sL |> perform_exp(design_sampling("group",.6)) |> scale_scans()

#> 
#> Scaled weighted adjacency matrix                               
#> a .     0.944 0.258 0.487 0.354
#> b 0.929 .     0.063 0.347 0.493
#> c 0.338 0.422 .     0.121 0.136
#> d 0.667 1.000 0.338 .     0.950
#> e 0.224 0.507 0.408 0.467 .    
#> 
#> 
#> Hidden attributes:
#> scanList.type - raw.scanList - Adj - samp.effort - n.scans - mode
#> Adj.subfun - edge.Prob - obs.P - theoretical.scanList - summed.scanList - sampled

## comparing group-scan and focal sampling
sL |> perform_exp(design_sampling("group",.6),
                  design_sampling("focal","even")
                  ) |> scale_scans()

#> [[1]]
#> 
#> Scaled weighted adjacency matrix                               
#> a .     0.973 0.387 0.562 0.488
#> b 0.921 .     0.071 0.405 0.464
#> c 0.293 0.431 .     0.110 0.183
#> d 0.580 1.000 0.379 .     0.958
#> e 0.254 0.466 0.395 0.447 .    
#> 
#> 
#> Hidden attributes:
#> scanList.type - raw.scanList - Adj - samp.effort - n.scans - mode
#> Adj.subfun - edge.Prob - obs.P - theoretical.scanList - summed.scanList - sampled
#> 
#> [[2]]
#> 
#> Scaled weighted adjacency matrix                               
#> a .     0.958 0.354 0.646 0.417
#> b 0.875 .     0.104 0.375 0.438
#> c 0.312 0.521 .     0.146 0.083
#> d 0.604 1.000 0.396 .     1.000
#> e 0.188 0.583 0.333 0.438 .    
#> 
#> 
#> Hidden attributes:
#> scanList.type - raw.scanList - Adj - samp.effort - n.scans - mode
#> Adj.subfun - edge.Prob - focalList - theoretical.scanList - summed.scanList - sampled
#> 
#> attr(,"class")
#> [1] "sLlist"