kbdm_sews: Indicator based on Kolmogorov Complexity

View source: R/task_kbdm.R

kbdm_sewsR Documentation

Indicator based on Kolmogorov Complexity

Description

Computes the Kolmogorov Complexity on a set of matrices, using the Block Decomposition Method.

Usage

kbdm_sews(mat, subsize = 3)

Arguments

mat

A logical matrix (TRUE/FALSE values) or a list of logical matrices

subsize

A submatrix size to carry out the Block Decomposition Method (must be between 1 and 3)

Details

be a useful indicator to anticipate transitions in model ecological systems (Dakos and Soler-Toscano, 2017). When close to the transition critical point, the complexity is expected to decrease.

The Kolmogorov complexity cannot be computed directly for large strings (i.e. matrices). However, the complexity of smaller submatrices can be estimated, then combined to obtain an approximation of the complexity of the whole matrix. This method, the Block Decomposition Method is implemented in this indicator following Dakos and Soler-Toscano (2017).

Value

kbdm_sews returns an object of class simple_sews_single (a list) if mat is a single matrix, and an object of class simple_sews_list if mat is a list of matrices. These objects can be used with generic methods indictest (to test significance) or plot (to display trends), see also the examples below.

References

Dakos, V., and F. Soler-Toscano. 2017. Measuring complexity to infer changes in the dynamics of ecological systems under stress. Ecological Complexity 32:144-155.

See Also

raw_kbdm, acss, indictest, to test the significance of indicator values.

Examples


 

kbdm_result <- kbdm_sews(serengeti, subsize = 3)
plot(kbdm_result, along = serengeti.rain)

kbdm_test <- indictest(kbdm_result, nulln = 49)
plot(kbdm_test, along = serengeti.rain)

# Plot deviation to null expectation
plot(kbdm_test, along = serengeti.rain, what = "z_score") 





spatialwarnings documentation built on Sept. 11, 2024, 8:55 p.m.