`R/PLNmixturefit-class.R`

`PLNmixturefit.Rd`

The function `PLNmixture`

produces a collection of models which are instances of object with class `PLNmixturefit`

.
A `PLNmixturefit`

(say, with k components) is itself a collection of k `PLNfit`

.

This class comes with a set of methods, some of them being useful for the user: See the documentation for ...

The function `PLNmixture`

, the class `PLNmixturefamily`

`n`

number of samples

`p`

number of dimensions of the latent space

`k`

number of components

`d`

number of covariates

`components`

components of the mixture (PLNfits)

`latent`

a matrix: values of the latent vector (Z in the model)

`latent_pos`

a matrix: values of the latent position vector (Z) without covariates effects or offset

`posteriorProb`

matrix ofposterior probability for cluster belonging

`memberships`

vector for cluster index

`mixtureParam`

vector of cluster proportions

`optim_par`

a list with parameters useful for monitoring the optimization

`nb_param`

number of parameters in the current PLN model

`entropy_clustering`

Entropy of the variational distribution of the cluster (multinomial)

`entropy_latent`

Entropy of the variational distribution of the latent vector (Gaussian)

`entropy`

Full entropy of the variational distribution (latent vector + clustering)

`loglik`

variational lower bound of the loglikelihood

`loglik_vec`

element-wise variational lower bound of the loglikelihood

`BIC`

variational lower bound of the BIC

`ICL`

variational lower bound of the ICL (include entropy of both the clustering and latent distributions)

`R_squared`

approximated goodness-of-fit criterion

`criteria`

a vector with loglik, BIC, ICL, and number of parameters

`model_par`

a list with the matrices of parameters found in the model (Theta, Sigma, Mu and Pi)

`vcov_model`

character: the model used for the covariance (either "spherical", "diagonal" or "full")

`fitted`

a matrix: fitted values of the observations (A in the model)

`group_means`

a matrix of group mean vectors in the latent space.

`new()`

Optimize a the

Initialize a `PLNmixturefit`

model

```
PLNmixturefit$new(
responses,
covariates,
offsets,
posteriorProb,
formula,
control
)
```

`responses`

the matrix of responses common to every models

`covariates`

the matrix of covariates common to every models

`offsets`

the matrix of offsets common to every models

`posteriorProb`

matrix ofposterior probability for cluster belonging

`formula`

model formula used for fitting, extracted from the formula in the upper-level call

`control`

a list for controlling the optimization.

`optimize()`

Optimize a `PLNmixturefit`

model

`predict()`

Predict group of new samples

```
PLNmixturefit$predict(
newdata,
type = c("posterior", "response", "position"),
prior = matrix(rep(1/self$k, self$k), nrow(newdata), self$k, byrow = TRUE),
control = PLNmixture_param(),
envir = parent.frame()
)
```

`newdata`

A data frame in which to look for variables, offsets and counts with which to predict.

`type`

The type of prediction required. The default

`posterior`

are posterior probabilities for each group ,`response`

is the group with maximal posterior probability and`latent`

is the averaged latent coordinate (without offset and nor covariate effects), with weights equal to the posterior probabilities.`prior`

User-specified prior group probabilities in the new data. The default uses a uniform prior.

`control`

a list-like structure for controlling the fit. See

`PLNmixture_param()`

for details.`envir`

Environment in which the prediction is evaluated

`plot_clustering_data()`

Plot the matrix of expected mean counts (without offsets, without covariate effects) reordered according the inferred clustering

```
PLNmixturefit$plot_clustering_data(
main = "Expected counts reorder by clustering",
plot = TRUE,
log_scale = TRUE
)
```

`plot_clustering_pca()`

Plot the individual map of a PCA performed on the latent coordinates, where individuals are colored according to the memberships

```
PLNmixturefit$plot_clustering_pca(
main = "Clustering labels in Individual Factor Map",
plot = TRUE
)
```

`postTreatment()`

Update fields after optimization

`responses`

the matrix of responses common to every models

`covariates`

the matrix of covariates common to every models

`offsets`

the matrix of offsets common to every models

`weights`

an optional vector of observation weights to be used in the fitting process.

`config`

a list for controlling the post-treatment

`nullModel`

null model used for approximate R2 computations. Defaults to a GLM model with same design matrix but not latent variable.