term is an R package to create, manipulate, query and repair vectors of parameter terms. Parameter terms are the labels used to reference values in vectors, matrices and arrays. They represent the names in coefficient tables and the column names in mcmc and mcmc.list objects.
To install the latest release from CRAN
install.packages("term")
To install the developmental version from GitHub
# install.packages("remotes") remotes::install_github("poissonconsulting/term")
library(term) # term vectors are the labels used to reference values in # vectors, matrices and arrays term <- term( "alpha[1]", "alpha[2]", "beta[1,1]", "beta[2,1]", "beta[1,2]", "beta[2,2]", "sigma" ) # they print like character vectors term #> <term[7]> #> [1] alpha[1] alpha[2] beta[1,1] beta[2,1] beta[1,2] beta[2,2] sigma # but are S3 class objects that also inherit from vctrs_vctr class(term) #> [1] "term" "vctrs_vctr"
# consider a matrix of term values set.seed(101) estimate <- matrix(rnorm(4), nrow = 2) estimate #> [,1] [,2] #> [1,] -0.3260365 -0.6749438 #> [2,] 0.5524619 0.2143595 # the term labels can be created using as_term() term <- as_term(estimate, name = "b0") term #> <term[4]> #> [1] b0[1,1] b0[2,1] b0[1,2] b0[2,2] # and combined with the term values to create a coef table library(tibble) coef <- tibble(term = term, estimate = as.vector(estimate)) coef #> # A tibble: 4 x 2 #> term estimate #> <term> <dbl> #> 1 b0[1,1] -0.326 #> 2 b0[2,1] 0.552 #> 3 b0[1,2] -0.675 #> 4 b0[2,2] 0.214
# the term vectors are readily sorted coef[rev(order(coef$term)),] #> # A tibble: 4 x 2 #> term estimate #> <term> <dbl> #> 1 b0[2,2] 0.214 #> 2 b0[1,2] -0.675 #> 3 b0[2,1] 0.552 #> 4 b0[1,1] -0.326 # and the parameter names or parameter dimensions extracted term <- term( "alpha[1]", "alpha[2]", "beta[1,1]", "beta[2,1]", "beta[1,2]", "beta[2,2]", "sigma" ) pars(term) #> [1] "alpha" "beta" "sigma" pdims(term) #> $alpha #> [1] 2 #> #> $beta #> [1] 2 2 #> #> $sigma #> [1] 1 # this can also be done for each term pars_terms(term) #> [1] "alpha" "alpha" "beta" "beta" "beta" "beta" "sigma" tindex(term) #> $`alpha[1]` #> [1] 1 #> #> $`alpha[2]` #> [1] 2 #> #> $`beta[1,1]` #> [1] 1 1 #> #> $`beta[2,1]` #> [1] 2 1 #> #> $`beta[1,2]` #> [1] 1 2 #> #> $`beta[2,2]` #> [1] 2 2 #> #> $sigma #> [1] 1
# term vectors can be tested for whether they have (parseably) valid, # (dimensionally) consistent and complete terms. # valid terms valid_term(term("a", "a[1]", "a [2]", " b [3 ] ", "c[1,10]")) #> [1] TRUE TRUE TRUE TRUE TRUE # invalid terms valid_term(new_term(c("a a", "a[]", "a[2", " b[3 3]", "c[1,10]c"))) #> [1] FALSE FALSE FALSE FALSE FALSE # consistent terms consistent_term(term("a", "a[2]", "b[1,1]", "b[10,99]")) #> [1] TRUE TRUE TRUE TRUE # inconsistent terms consistent_term(term("a", "a[2,1]", "b[1,1]", "b[10,99,1]")) #> [1] FALSE FALSE FALSE FALSE # complete terms is_incomplete_terms(term("a", "a[2]", "b[1,1]", "b[2,1]")) #> [1] FALSE # incomplete terms is_incomplete_terms(term("a", "a[3]", "b[1,1]", "b[2,2]")) #> [1] TRUE
term <- new_term(c("b[4]", "b [2]", "b", "b[1", "b[2, 2]", "b", "a [ 1 ] ", NA)) term #> <term[8]> #> [1] b[4] `b [2]` b b[1 `b[2, 2]` b `a [ 1 ] ` #> [8] <NA> # valid terms can be repaired (invalid terms are converted to missing values) term <- repair_terms(term) term #> <term[8]> #> [1] b[4] b[2] b[1] <NA> b[2,2] b[1] a <NA> # the `term()` constructor rejects invalid terms term("b[4]", "b [2]", "b", "b[1", "b[2, 2]", "b", "a [ 1 ] ", NA) #> Error: All elements of term vector `string_args_term` must be valid. # missing values can easily removed term <- term[!is.na(term)] term #> <term[6]> #> [1] b[4] b[2] b[1] b[2,2] b[1] a # and only unique values retained term <- unique(term) term #> <term[5]> #> [1] b[4] b[2] b[1] b[2,2] a # a term vector can be sorted by parameter name and index term <- sort(term) term #> <term[5]> #> [1] a b[1] b[2] b[4] b[2,2] # an inconsistent term removed term <- term[term != "b[2,2]"] term #> <term[4]> #> [1] a b[1] b[2] b[4] # and incomplete terms completed term <- sort(complete_terms(term)) term #> <term[5]> #> [1] a b[1] b[2] b[3] b[4]
Please note that the term project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.