The goal of sinar is to implement the Conditional Least Square method for the Spatial non-negative Integer-valued Autoregressive (SINAR(1,1)).
You can install the development version from GitHub with:
# install.packages("devtools")
devtools::install_github("gilberto-sassi/sinar")
library(sinar)
## Simulated data matrix from SINAR(1,1) with Poison(5) innovation
matrix_simulated <- sinar_pois(15, 15, 0.2, 0.2, 0.4, 5)
## Conditional Least Square (CLS) estimates
cls(matrix_simulated)
#> a10 a01 a11 mu
#> 0.1605389 0.2860054 0.4277413 3.1261927
## Covariance matrix of CLS estimates
emp_cov(matrix_simulated)
#> a10 a01 a11 mu
#> a10 0.0044018403 0.0001991086 -0.001362643 -0.08051497
#> a01 0.0001991086 0.0032884060 -0.000882474 -0.06218858
#> a11 -0.0013626431 -0.0008824740 0.004125110 -0.04507648
#> mu -0.0805149667 -0.0621885767 -0.045076478 4.67716808
library(sinar)
## Nematodes counting datasets
data("nematodes")
## Conditional Least Square (CLS) estimates
cls(nematodes)
#> a10 a01 a11 mu
#> 0.20664577 0.33147378 0.04523086 2.14476453
## Covariance matrix of CLS estimates
emp_cov(nematodes)
#> a10 a01 a11 mu
#> a10 0.0111169222 -0.0009999304 -0.003310576 -0.017278481
#> a01 -0.0009999304 0.0082946407 -0.001503724 -0.009838536
#> a11 -0.0033105760 -0.0015037242 0.004507501 0.004049939
#> mu -0.0172784806 -0.0098385364 0.004049939 0.268045835
library(sinar)
## Carabidae counting dataset
data("carabidae")
## Conditional Least Square (CLS) estimates
cls(carabidae)
#> a10 a01 a11 mu
#> 0.14595392 0.12725313 0.08798513 9.10361759
## Covariance matrix of CLS estimates
emp_cov(carabidae)
#> a10 a01 a11 mu
#> a10 0.014484776 -0.003141815 -0.005525906 -0.06795645
#> a01 -0.003141815 0.014365625 -0.001265544 -0.11558802
#> a11 -0.005525906 -0.001265544 0.023795735 -0.25417404
#> mu -0.067956449 -0.115588024 -0.254174036 7.22525572