Take one fitted model out of an enumeration grid or a comparison
Source:R/enumeration.R
get_candidate.RdReturns one fitted model from an enumeration, selected by its name or by its number of profiles and group classes (and its covariance structure when the grid has several), or one model from a comparison, selected by its row label.
Usage
get_candidate(x, n_profiles, n_group_classes = 1L, model = NULL)
candidate_fit(x, n_profiles, n_group_classes = 1L, model = NULL)Arguments
- x
An enumeration grid from
enumerate_classes()(a profile, latent class, transition or group-class family grid), or a comparison fromcompare_models(). On a family grid, give the name, orn_group_classeswithmodel.- n_profiles
Number of individual profiles of the candidate, the candidate's name (
"profiles3_groups4"), or its two class counts (c(3, 4)). For a comparison, the row label, as itsmodelcolumn shows it.- n_group_classes
Number of group classes of the candidate. Default 1.
- model
The covariance structure of the candidate, needed when
enumerate_classes()crossed several structures and the class counts match several candidates; on a group-class family grid, the family model as the grid'smodelcolumn shows it.
Value
The fitted model of that candidate, the same kind of object its
fitting function returns (for a profile grid, class multilpa), so the
verbs for a fitted model apply to it.
Details
A candidate is selected by its name, as the grid's candidate column shows
it (get_candidate(grid, "profiles3_groups4"); classes instead of
profiles when every indicator is categorical); by its class counts,
get_candidate(grid, 3, 4) or get_candidate(grid, c(3, 4)); or, on a grid
that crosses covariance structures, by its counts and model. From a
comparison, give the row label: get_candidate(comparison, "profiles3_groups4").
candidate_fit() is the former name of get_candidate(), kept
for compatibility. It takes the same arguments, forwards to
get_candidate() and gives no warning.
Conditions
latents_unknown_candidate for a name or class counts not in the grid,
for counts that match several candidates (the message lists their
names), and for a label that is not a row of a comparison;
latents_failed_candidate when the candidate's fit raised an error, so
there is no model to return; latents_bad_argument for a name or pair of
counts given together with n_group_classes or model, for a value that
names no candidate, and for a comparison given no label;
latents_incomplete_fit for a comparison that no longer carries its
models. Counts that are not single positive whole numbers fail the
argument checks.
Examples
set.seed(1)
d <- data.frame(g = rep(1:10, each = 10), y = rnorm(100))
candidates <- enumerate_classes(d, "y", "g", n_profiles = 1:2,
n_group_classes = 1, n_starts = 2, seed = 1)
get_candidate(candidates, n_profiles = 2, n_group_classes = 1, model = "VVI")
#> Two-level latent profile analysis: 2 profiles, 1 group class
#> 100 individuals in 10 groups; varying diagonal residual covariance (VVI)
#> Log likelihood: -130.591962 | AIC: 271.184 | BIC (groups): 272.697
#> Converged: TRUE | iterations: 36 | best start: 1/2
#>
#> profile y count proportion
#> 1 -0.1928460 44.5831 0.445831
#> 2 0.3516482 55.4169 0.554169
#>
#> Variances and standard errors: get_results(x, "profiles").
#> Every other table: get_results(x, what = ), or get_results(x, "all").
# This grid crosses two structures, so the names carry them:
candidates
#> Class enumeration: 4 candidate models
#> candidate n_profiles n_group_classes model log_likelihood
#> profiles1_groups1_EEI 1 1 EEI -130.6550
#> profiles2_groups1_EEI 2 1 EEI -130.5791
#> profiles1_groups1_VVI 1 1 VVI -130.6550
#> profiles2_groups1_VVI 2 1 VVI -130.5920
#> n_parameters aic bic_groups bic_individual icl_individual profile_entropy
#> 2 265.3100 265.9152 270.5204 270.5204 NA
#> 4 269.1581 270.3684 279.5788 355.6432 0.45131114
#> 2 265.3100 265.9152 270.5204 270.5204 NA
#> 5 271.1839 272.6968 284.2098 411.5897 0.08114782
#> group_entropy avepp_min avepp_max avepp_mean avepp_below group_avepp_min
#> NA NA NA NA NA NA
#> NA 0.6413011 0.8527078 0.7470045 1 NA
#> NA NA NA NA NA NA
#> NA 0.6380900 0.6405463 0.6393181 2 NA
#> group_avepp_max group_avepp_mean group_avepp_below converged boundary
#> NA NA NA TRUE FALSE
#> NA NA NA TRUE FALSE
#> NA NA NA TRUE FALSE
#> NA NA NA TRUE FALSE
#> n_best_replicated
#> 2
#> 2
#> 2
#> 2
#> No candidate is selected automatically. Compare one convention consistently.
get_candidate(candidates, "profiles2_groups1_VVI")
#> Two-level latent profile analysis: 2 profiles, 1 group class
#> 100 individuals in 10 groups; varying diagonal residual covariance (VVI)
#> Log likelihood: -130.591962 | AIC: 271.184 | BIC (groups): 272.697
#> Converged: TRUE | iterations: 36 | best start: 1/2
#>
#> profile y count proportion
#> 1 -0.1928460 44.5831 0.445831
#> 2 0.3516482 55.4169 0.554169
#>
#> Variances and standard errors: get_results(x, "profiles").
#> Every other table: get_results(x, what = ), or get_results(x, "all").
compared <- compare_models(candidates, "profiles1_groups1_VVI",
"profiles2_groups1_VVI")
get_candidate(compared, "profiles2_groups1_VVI")
#> Two-level latent profile analysis: 2 profiles, 1 group class
#> 100 individuals in 10 groups; varying diagonal residual covariance (VVI)
#> Log likelihood: -130.591962 | AIC: 271.184 | BIC (groups): 272.697
#> Converged: TRUE | iterations: 36 | best start: 1/2
#>
#> profile y count proportion
#> 1 -0.1928460 44.5831 0.445831
#> 2 0.3516482 55.4169 0.554169
#>
#> Variances and standard errors: get_results(x, "profiles").
#> Every other table: get_results(x, what = ), or get_results(x, "all").