### Install and Load remstimate Package

Source: https://tilburgnetworkgroup.github.io/remstimate/index.html

Installs the remstimate package from CRAN and loads it into the R session. This is the initial setup required to use the package's functionalities.

```r
install.packages("remstimate")
library(remify)
```

--------------------------------

### Example: Tie-oriented model MLE diagnostics

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/plot.remstimate.html

Demonstrates the setup for plotting diagnostics of a tie-oriented model using the Maximum Likelihood Estimation (MLE) approach. This involves loading data, processing it with 'remify', specifying the model, and calculating statistics.

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh, 
                                    tie_effects = tie_model)

```

--------------------------------

### Print remstimate Object (Tie-Oriented Model)

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/print.remstimate.html

This example demonstrates how to print a remstimate object for a tie-oriented model. It includes data loading, model specification, statistics calculation, and estimation.

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh,
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "BSIR",
                                  nsim = 100,
                                  ncores = 1)

# print
tie_mle
```

--------------------------------

### Estimate Tie-Oriented Model with remstimate

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/bic.html

This example shows the full process of estimating a tie-oriented model using remify, remstats, and remstimate. It includes data loading, processing, specifying the model, calculating statistics, and running the estimation.

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh,
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "MLE",
                                  ncores = 1)
```

--------------------------------

### Estimate Model Parameters with remstimate()

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Run the model estimation using the remstimate() function with specified statistics and method. This example uses the 'MLE' method.

```r
# for example the method "MLE"
remstimate::remstimate(reh = ao_reh,
                          stats =  ao_stats,
                          method = "MLE",
                          ncores = ncores)
```

--------------------------------

### Print remstimate Object (Actor-Oriented Model)

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/print.remstimate.html

This example shows how to print a remstimate object for an actor-oriented model. It covers data preparation, model specification for sender rate and receiver choice, statistics calculation, and estimation.

```R
# loading data
data(ao_data)

# processing event sequence with remify
ao_reh <- remify::remify(edgelist = ao_data$edgelist, model = "actor")
  
# specifying linear predictor (for sender rate and receiver choice model)
rate_model <- ~ 1 + remstats::indegreeSender()
choice_model <- ~ remstats::inertia() + remstats::reciprocity()

# calculating statistics
ao_reh_stats <- remstats::remstats(reh = ao_reh,
                                   sender_effects = rate_model,
                                   receiver_effects = choice_model)

# running estimation
ao_mle <- remstimate::remstimate(reh = ao_reh,
                                 stats = ao_reh_stats,
                                 method = "BSIR",
                                 nsim = 100,
                                 ncores = 1)

# print
ao_mle
```

--------------------------------

### List Available Vignettes

Source: https://tilburgnetworkgroup.github.io/remstimate/index.html

Lists all vignettes available for the remstimate package. Vignettes provide detailed examples and explanations of the package's features.

```r
vignette(package = "remstimate")
```

--------------------------------

### Estimate REM using BSIR with Custom Priors

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Fit a relational event model using the Bayesian Sampling Importance Resampling (BSIR) method. This example demonstrates setting custom prior distributions and their arguments for both sender and receiver models.

```R
library(mvnfast) # loading package for fast simulation from a multivariate Student t distribution
priormvt <- mvnfast::dmvt # defining which distribution we want to use from the 'mvnfast' package
ao_bsir <- remstimate::remstimate(reh = ao_reh,
                        stats =  ao_stats,
                        ncores = ncores,
                        method = "BSIR",
                        nsim = 100L, # 100 draws from the posterior distribution
                        prior = list(sender_model = priormvt, receiver_model = priormvt), #  defining prior here, prior parameters follow below
                        prior_args = list(sender_model =  list(mu = rep(0,dim(ao_stats$sender_stats)[3]), # prior mu value for sender_model
                                                            sigma = diag(dim(ao_stats$sender_stats)[3])*1.5, # prior sigma value for sender_model
                                                            df = 1),  # prior df value
                                        receiver_model = list(mu = rep(0,dim(ao_stats$receiver_stats)[3]), # prior mu value for receiver_model
                                                            sigma = diag(dim(ao_stats$receiver_stats)[3])*1.5, # prior sigma value for receiver_model
                                                            df = 1)), # prior df value
                        log = TRUE, # requiring log density values from the prior,
                        seed = 20929, # set a seed only for reproducibility purposes
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100 # number of draws for the computation of the WAIC set to 100     
                        )

# summary 
summary(ao_bsir)
```

--------------------------------

### Calculate AICC for a remstimate object

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/aicc.html

This example demonstrates how to calculate the AICC value for a 'remstimate' object after processing network data and running an estimation. Ensure remify, remstats, and remstimate packages are loaded and data is prepared.

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh, 
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "MLE",
                                  ncores = 1)

# AICC
aicc(tie_mle) 
#> [1] 1219.046
```

--------------------------------

### WAIC Function Signature

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/waic.html

Defines the S3 method signature for the waic function when applied to a 'remstimate' object. No examples are currently available.

```R
waic(object, ...)

# S3 method for class 'remstimate'
waic(object, ...)
```

--------------------------------

### Summary for tie-oriented model (BSIR)

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/summary.remstimate.html

Generates a summary for a tie-oriented relational event model estimated using the BSIR method. Requires prior estimation of the model.

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh,
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "BSIR",
                                  nsim = 100,
                                  ncores = 1)

# summary
summary(tie_mle)
```

--------------------------------

### Compute Diagnostics for Tie-Oriented Model (MLE)

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/diagnostics.html

This snippet demonstrates how to compute diagnostics for a tie-oriented model estimated using the Maximum Likelihood Estimation (MLE) method. It includes data loading, event sequence processing, specifying the linear predictor, calculating statistics, running the estimation, and finally computing the diagnostics.

```R
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh,
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "MLE",
                                  ncores = 1)

# diagnostics
tie_diagnostics <- diagnostics(object = tie_mle, reh = tie_reh, stats = tie_reh_stats)
names(tie_diagnostics)
#> [1] "residuals"      "rates"          ".reh.processed"
```

--------------------------------

### Estimate Model using MLE and Summarize

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/ao_data.html

Estimates a relational event model using the Maximum Likelihood Estimation (MLE) method with the 'remstimate' function and then prints a summary of the estimation results.

```r
mle_ao <- remstimate::remstimate(reh = ao_reh, stats = ao_reh_stats, method = "MLE")
summary(mle_ao)
```

--------------------------------

### Load remestimate Package in R

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Load the remestimate package to begin using its functionalities for relational event modeling. This also loads the 'remify' and 'remstats' packages.

```r
library(remstimate)
```

--------------------------------

### Summarize MLE estimation for tie-oriented model

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/tie_data.html

Prints a summary of the Maximum Likelihood Estimation (MLE) results for the tie-oriented relational event model, including coefficients, standard errors, and model fit statistics.

```r
summary(mle_tie)
```

--------------------------------

### Summarize remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Generate a detailed summary of the fitted relational event model, including model calls, coefficients with standard errors, z-values, p-values, and various model fit statistics.

```r
# summary of the 'remstimate' object
summary(ao_mle)
```

--------------------------------

### Summary for actor-oriented model (BSIR)

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/summary.remstimate.html

Generates a summary for an actor-oriented relational event model estimated using the BSIR method. This involves separate models for call rate and receiver choice.

```R
# loading data
data(ao_data)

# processing event sequence with remify
ao_reh <- remify::remify(edgelist = ao_data$edgelist, model = "actor")
  
# specifying linear predictor (for sender rate and receiver choice model)
rate_model <- ~ 1 + remstats::indegreeSender()
choice_model <- ~ remstats::inertia() + remstats::reciprocity()

# calculating statistics
ao_reh_stats <- remstats::remstats(reh = ao_reh,
                                   sender_effects = rate_model,
                                   receiver_effects = choice_model)

# running estimation
ao_mle <- remstimate::remstimate(reh = ao_reh,
                                 stats = ao_reh_stats,
                                 method = "BSIR",
                                 nsim = 100,
                                 ncores = 1)

# summary
summary(ao_mle)
```

--------------------------------

### Print remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Display the results of a fitted relational event model. This provides a summary of coefficients, deviance, and information criteria.

```r
# printing the 'remstimate' object 
ao_mle
```

--------------------------------

### Estimate Model with MLE and WAIC using remstimate

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Fit a relational event model using MLE and simultaneously compute the Widely Applicable Information Criterion (WAIC). This involves setting `WAIC = TRUE` and specifying `nsimWAIC` for the number of draws.

```R
tie_mle <- remstimate::remstimate(
                          reh = tie_reh,
                          stats = tie_stats,
                          ncores = ncores,
                          method = "MLE",
                          WAIC = TRUE, # setting WAIC computation to TRUE
                          nsimWAIC = 100) # number of draws for the computation of the WAIC set to 100         
```

--------------------------------

### Load tie_data

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/tie_data.html

Loads the tie_data object into the R workspace. This dataset contains a simulated sequence of relational events.

```r
data(tie_data)
```

--------------------------------

### Print remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Display the estimated coefficients and model fit statistics for a `remstimate` object. This provides a concise overview of the model results.

```R
# printing the 'remstimate' object
tie_mle
```

--------------------------------

### Plot RemEstimate Results

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/plot.remstimate.html

Creates a plot visualizing the estimation results and diagnostics. Requires the estimated object, reh, and diagnostics.

```R
plot(x = tie_mle, reh  = tie_reh, diagnostics = tie_diagnostics)
```

--------------------------------

### Load and Process Relational Event Data in R

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Loads simulated relational event data and processes it for actor-oriented modeling using the remify package. Sets the number of cores for processing.

```r
ncores <- 1L

# loading data
data(ao_data)

# true parameters' values
ao_data$true.pars
```

```r
# processing event sequence with 'remify'
ao_reh <- remify::remify(edgelist = ao_data$edgelist, model = "actor")

# summary of the relational event network
summary(ao_reh)
```

--------------------------------

### Set ncores and load tie_data

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Sets the number of cores for processing and loads the simulated relational event sequence data. Adjust `ncores` as needed.

```r
# setting `ncores` to 1 (the user can change this parameter)
ncores <- 1L

# loading data
data(tie_data)

# true parameters' values
tie_data$true.pars
```

--------------------------------

### Summarize remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Generate a detailed summary of the `remstimate` object, including coefficient estimates, standard errors, z-values, p-values, and model fit information. This is useful for in-depth analysis of the model.

```R
# summary of the 'remstimate' object
summary(tie_mle)
```

--------------------------------

### Run RemEstimate MLE

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/plot.remstimate.html

Runs a Maximum Likelihood Estimation (MLE) for network data. Requires reh and stats objects. Specify ncores for parallel processing.

```R
tie_mle <- remstimate::remstimate(
  reh = tie_reh,
  stats = tie_reh_stats,
  method = "MLE",
  ncores = 1
)
```

--------------------------------

### Estimate Model using MLE with remstimate

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Estimate the relational event model using the Maximum Likelihood Estimation (MLE) method with the `remstimate` function. This requires the processed network, calculated statistics, and the desired method.

```R
# for example the method "MLE"
remstimate::remstimate(reh = tie_reh,
                          stats =  tie_stats,
                          method = "MLE",
                          ncores = ncores)    
```

--------------------------------

### Perform Diagnostics

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/plot.remstimate.html

Generates diagnostic statistics for a remestimate object. Requires the estimated object, reh, and stats.

```R
tie_diagnostics <- diagnostics(object = tie_mle, reh = tie_reh, stats = tie_reh_stats)
```

--------------------------------

### Process event sequence with remify

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/tie_data.html

Processes the raw edgelist from tie_data into a relational event history object using the remify function with a tie-oriented model.

```r
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
```

--------------------------------

### Estimate tie-oriented model using MLE

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/tie_data.html

Estimates the parameters of the tie-oriented relational event model using the Maximum Likelihood Estimation (MLE) method with the remstimate function. Requires the remified event history and calculated statistics.

```r
mle_tie <- remstimate::remstimate(reh = tie_reh, stats = tie_reh_stats, method = "MLE")
```

--------------------------------

### Estimate Model with MLE and WAIC Calculation

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Perform Maximum Likelihood Estimation (MLE) for an actor-oriented model, including the computation of WAIC. Set WAIC computation to TRUE and specify the number of draws for WAIC calculation.

```r
ao_mle <- remstimate::remstimate(reh = ao_reh,
                        stats = ao_stats,
                        ncores = ncores,
                        method = "MLE",
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100) # number of draws for the computation of the WAIC set to 100
```

--------------------------------

### Calculate Diagnostics for remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Compute diagnostic statistics for a `remstimate` model object. This requires the fitted model, the original relational event network, and the calculated statistics.

```R
# diagnostics
tie_mle_diagnostics <- diagnostics(object = tie_mle, reh = tie_reh, stats = tie_stats)
```

--------------------------------

### Load ao_data Dataset

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/ao_data.html

Loads the predefined 'ao_data' dataset into the R workspace. This dataset contains a simulated sequence of relational events.

```r
data(ao_data)
```

--------------------------------

### R S3 method for remstimate object

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/plot.remstimate.html

This is the S3 method signature for plotting diagnostics of a 'remstimate' object. It outlines the main parameters for controlling which diagnostics are plotted and for which effects.

```R
# S3 method for class 'remstimate'
plot(
  x,
  reh,
  diagnostics = NULL,
  which = c(1:4),
  effects = NULL,
  sender_effects = NULL,
  receiver_effects = NULL,
  ...
)
```

--------------------------------

### Open Specific Vignette

Source: https://tilburgnetworkgroup.github.io/remstimate/index.html

Opens a specific vignette for the remstimate package, such as the main 'remstimate' vignette. This is useful for understanding how to use the package in practice.

```r
vignette(topic = "remstimate", package = "remstimate") # or simply vignette("remstimate")
```

--------------------------------

### Run HMC Diagnostics

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Runs Hamiltonian Monte Carlo (HMC) diagnostics on a tie object. This is useful for assessing the convergence and quality of MCMC simulations.

```R
tie_hmc_diagnostics <- diagnostics(object = tie_hmc, reh = tie_reh, stats = tie_stats)
```

--------------------------------

### Estimate Actor-Oriented Model using MLE

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/remstimate.html

Estimates parameters for an actor-oriented relational event model using Maximum Likelihood Estimation (MLE). This involves specifying separate models for sender rate and receiver choice, along with processed event data.

```R
# loading data
data(ao_data)

# processing event sequence with remify
ao_reh <- remify::remify(edgelist = ao_data$edgelist, model = "actor")
  
# specifying linear predictor (for sender rate and receiver choice model)
rate_model <- ~ 1 + remstats::indegreeSender()
choice_model <- ~ remstats::inertia() + remstats::reciprocity()

# calculating statistics
ao_reh_stats <- remstats::remstats(reh = ao_reh,
                                   sender_effects = rate_model,
                                   receiver_effects = choice_model)

# running estimation
ao_mle <- remstimate::remstimate(reh = ao_reh,
                                 stats = ao_reh_stats,
                                 method = "MLE",
                                 ncores = 1)
# summary
summary(ao_mle)
```

--------------------------------

### Estimate REM with BSIR

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Employ Bayesian Sampling Importance Resampling (BSIR) for estimation. Requires loading the 'mvnfast' package and specifying prior distributions and parameters. WAIC computation is optional.

```r
library(mvnfast) # loading package for fast simulation from a multivariate Student t distribution
priormvt <- mvnfast::dmvt # defining which distribution we want to use from the 'mvnfast' package
tie_bsir <- remstimate::remstimate(reh = tie_reh,
                        stats =  tie_stats,
                        ncores = ncores,
                        method = "BSIR",
                        nsim = 200L, # 200 draws from the posterior distribution
                        prior = priormvt, # defining prior here, prior parameters follow below
                        mu = rep(0,dim(tie_stats)[3]), # prior mu value
                        sigma = diag(dim(tie_stats)[3])*1.5, # prior sigma value
                        df = 1, # prior df value
                        log = TRUE, # requiring log density values from the prior,
                        seed = 23029, # set a seed only for reproducibility purposes
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100 # number of draws for the computation of the WAIC set to 100
                        )

# summary
summary(tie_bsir)
```

--------------------------------

### Calculate Network Statistics with remstats()

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Compute the network statistics defined in the linear predictor using the remstats() function. This prepares the data for model estimation.

```r
# calculating statistics (with `remstats`)
ao_stats <- remstats::remstats(reh = ao_reh, sender_effects = rate_model, receiver_effects = choice_model)

# the 'aomstats' 'remstats' object
ao_stats
```

--------------------------------

### Calculate statistics for tie-oriented model

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/tie_data.html

Calculates the necessary statistics for the defined tie-oriented model using the remstats function, based on the remified event history.

```r
tie_reh_stats <- remstats::remstats(reh = tie_reh, tie_effects = tie_model)
```

--------------------------------

### Estimate REM using GDADAMAX

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Fit a relational event model using the Adaptive Gradient Descent Optimization (GDADAMAX) method. This method requires specifying the number of epochs and optionally WAIC computation.

```R
ao_gd <- remstimate::remstimate(reh = ao_reh,
                        stats =  ao_stats,
                        ncores = ncores,
                        method = "GDADAMAX",
                        epochs = 200L, # number of iterations of the Gradient-Descent algorithm
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100) # number of draws for the computation of the WAIC set to 100     
# print 
ao_gd
```

--------------------------------

### Plot remstimate Object with Diagnostics

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Generate a plot visualizing the results of a `remstimate` model, incorporating diagnostic information. This helps in assessing the model's fit and performance.

```R
# plot
plot(x = tie_mle, reh  = tie_reh, diagnostics = tie_mle_diagnostics)
```

--------------------------------

### Specify Linear Predictor with remstats

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Define the linear predictor for the relational event model using a formula interface with `remstats`. This allows for the inclusion of various network statistics.

```R
# specifying linear predictor (with `remstats`) using a 'formula'
tie_model <- ~ 1 + remstats::indegreeSender() + 
              remstats::inertia() + remstats::reciprocity() 
```

--------------------------------

### Define Linear Predictors and Calculate Statistics

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/ao_data.html

Defines the linear predictors for both the rate and choice models using 'remstats' functions and then calculates the corresponding statistics for the relational event history.

```r
rate_model <- ~ 1 + remstats::indegreeSender()
choice_model <- ~ remstats::inertia() + remstats::reciprocity()
ao_reh_stats <- remstats::remstats(reh = ao_reh, sender_effects = rate_model, 
receiver_effects = choice_model)
```

--------------------------------

### Estimate REM with GDADAMAX

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Use GDADAMAX for adaptive gradient descent optimization. Set WAIC computation to TRUE and specify the number of draws for WAIC calculation.

```r
tie_gd <- remstimate::remstimate(reh = tie_reh,
                        stats =  tie_stats,
                        ncores = ncores,
                        method = "GDADAMAX",
                        epochs = 200L, # number of iterations for the Gradient-Descent algorithm
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100) # number of draws for the computation of the WAIC set to 100
# print
tie_gd
```

--------------------------------

### BibTeX Citation for remstimate Package

Source: https://tilburgnetworkgroup.github.io/remstimate/authors.html

Use this BibTeX entry for academic citations of the remstimate R package. Ensure the version and URL are up-to-date.

```bibtex
@Manual{
  title = {remstimate: Optimization Frameworks for Tie-Oriented and Actor-Oriented Relational Event Models},
  author = {Giuseppe Arena and Rumana Lakdawala and Fabio {Generoso Vieira}},
  year = {2025},
  note = {R package version 2.3.14},
  url = {https://tilburgnetworkgroup.github.io/remstimate/},
}
```

--------------------------------

### Compute Diagnostics

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/diagnostics.html

The `diagnostics` function computes and returns the diagnostics of a `remstimate` object. This includes standardized Schoenfeld's residuals and estimated event rates.

```APIDOC
## diagnostics

### Description
Computes the diagnostics of a `remstimate` object. The output object contains the residuals of the model estimated in the `remstimate` object, and the event rates estimated from the model at each time point.

### Method Signature
```R
diagnostics(object, reh, stats, ...)

# S3 method for class 'remstimate'
diagnostics(object, reh, stats, ...)
```

### Arguments
* `object` (remstimate): The `remstimate` object for which to compute diagnostics.
* `reh` (remify): The `remify` object used for the `remstimate` object.
* `stats` (remstats): The `remstats` object used for the `remstimate` object.
* `...`: Further arguments to be passed to the `diagnostics` method.

### Value
A object of class `"diagnostics" "remstimate"` containing standardized Schoenfeld's residuals and estimated event rates.

### Examples
```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")

# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh,
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "MLE",
                                  ncores = 1)

# diagnostics
tie_diagnostics <- diagnostics(object = tie_mle, reh = tie_reh, stats = tie_reh_stats)
names(tie_diagnostics)
#> [1] "residuals"      "rates"          ".reh.processed"
```
```

--------------------------------

### remstimate Function Signature

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/remstimate.html

This is the usage signature for the remstimate function. It outlines the required and optional parameters for optimizing network likelihoods.

```R
remstimate(
  reh,
  stats,
  method = c("MLE", "GDADAMAX", "BSIR", "HMC"),
  ncores = attr(reh, "ncores"),
  prior = NULL,
  nsim = 1000L,
  nchains = 1L,
  burnin = 500L,
  thin = 10L,
  init = NULL,
  epochs = 1000L,
  L = 50L,
  epsilon = ifelse(method == "GDADAMAX", 0.001, 0.002),
  seed = NULL,
  WAIC = FALSE,
  silent = TRUE,
  ...
)
```

--------------------------------

### Process Event Sequence with remify

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/ao_data.html

Processes the raw event sequence from 'ao_data$edgelist' using the 'remify' function to create a relational event history object suitable for actor-oriented models.

```r
ao_reh <- remify::remify(edgelist = ao_data$edgelist, model = "actor")
```

--------------------------------

### Estimate REM using Hamiltonian Monte Carlo (HMC)

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Estimates a Relational Event Model using the Hamiltonian Monte Carlo (HMC) method. This function requires specifying various simulation parameters such as the number of cores, simulation draws, chains, burn-in, leapfrog steps, step size, and thinning. WAIC computation can be enabled.

```R
ao_hmc <- remstimate::remstimate(reh = ao_reh,
                        stats =  ao_stats,
                        method = "HMC",
                        ncores = ncores,
                        nsim = 300L, # 300 draws to generate per each chain
                        nchains = 4L, # 4 chains (each one long 200 draws) to generate
                        burnin = 300L, # burnin length is 300
                        L = 100L, # number of leap-frog steps
                        epsilon = 0.1/100, # size of a leap-frog step
                        thin = 2L, # thinning size (this will reduce the final length of each chain will be 150)
                        seed = 23029, # set a seed only for reproducibility purposes
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100 # number of draws for the computation of the WAIC set to 100     
                        )
```

--------------------------------

### Generate Diagnostics for REM Estimate

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Generates diagnostic statistics for a fitted REM object. This is useful for assessing the convergence and quality of the model estimation.

```R
ao_bsir_diagnostics <- diagnostics(object = ao_bsir, reh = ao_reh, stats = ao_stats)
```

```R
ao_hmc_diagnostics <- diagnostics(object = ao_hmc, reh = ao_reh, stats = ao_stats)
```

--------------------------------

### remstimate Function

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/remstimate.html

Optimizes tie-oriented and actor-oriented likelihood. Supports MLE, GDADAMAX, BSIR, and HMC methods.

```APIDOC
## remstimate

### Description
A function for the optimization of tie-oriented and actor-oriented likelihood. There are four optimization algorithms: two Frequentists, Maximum Likelihood Estimation (`MLE`) and Adaptive Gradient Descent (`GDADAMAX`), and two Bayesian, Bayesian Sampling Importance Resampling (`BSIR`) and Hamiltonian Monte Carlo (`HMC`).

### Usage
```R
remstimate(
  reh,
  stats,
  method = c("MLE", "GDADAMAX", "BSIR", "HMC"),
  ncores = attr(reh, "ncores"),
  prior = NULL,
  nsim = 1000L,
  nchains = 1L,
  burnin = 500L,
  thin = 10L,
  init = NULL,
  epochs = 1000L,
  L = 50L,
  epsilon = ifelse(method == "GDADAMAX", 0.001, 0.002),
  seed = NULL,
  WAIC = FALSE,
  silent = TRUE,
  ...
)
```

### Parameters
* **reh**: Input object for the estimation.
* **stats**: Statistical data for the estimation.
* **method**: Optimization algorithm to use. Options include "MLE", "GDADAMAX", "BSIR", "HMC". Defaults to "MLE".
* **ncores**: Number of cores to use for computation. Defaults to the value in the `reh` attribute.
* **prior**: Prior distribution information (for Bayesian methods).
* **nsim**: Number of simulations (for BSIR).
* **nchains**: Number of MCMC chains (for HMC).
* **burnin**: Number of burn-in iterations (for HMC).
* **thin**: Thinning interval for MCMC samples (for HMC).
* **init**: Initial values for the optimization.
* **epochs**: Number of training epochs (for GDADAMAX).
* **L**: Number of leapfrog steps (for HMC).
* **epsilon**: Step size for gradient descent (GDADAMAX) or HMC.
* **seed**: Random seed for reproducibility.
* **WAIC**: Whether to compute Widely Applicable Information Criterion.
* **silent**: Suppress output messages.
* **...**: Additional arguments.
```

--------------------------------

### Define Linear Predictor for Rate and Choice Models

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Specify the linear predictor for actor-oriented rate and choice models using remstats functions. This sets up the effects to be estimated.

```r
# specifying linear predictor (for rate and choice model, with `remstats`)
rate_model <- ~ 1 + remstats::indegreeSender()
choice_model <- ~ remstats::inertia() + remstats::reciprocity()
```

--------------------------------

### Plot REM Model Results

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Visualize the results of a fitted REM model, including diagnostics. Ensure the model object, event data, and diagnostics object are available.

```R
plot(x = ao_mle, reh  = ao_reh, diagnostics = ao_mle_diagnostics)
```

--------------------------------

### Define linear predictor for tie-oriented model

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/tie_data.html

Defines the linear predictor for a tie-oriented relational event model, including terms for baseline, indegree of the sender, inertia, and reciprocity.

```r
tie_model <- ~ 1 + remstats::indegreeSender() + remstats::inertia() + remstats::reciprocity()
```

--------------------------------

### Estimate REM with HMC

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Utilize Hamiltonian Monte Carlo (HMC) for estimation. Specify the number of simulations per chain, number of chains, burn-in period, and thinning interval. WAIC computation is optional.

```r
tie_hmc <- remstimate::remstimate(reh = tie_reh,
                        stats =  tie_stats,
                        method = "HMC",
                        ncores = ncores,
                        nsim = 200L, # 200 draws to generate per each chain
                        nchains = 4L, # 4 chains to generate
                        burnin = 200L, # burnin length is 200
                        thin = 2L, # thinning size set to 2 (the final length of the chains will be 100)
                        seed = 23029, # set a seed only for reproducibility purposes
                        WAIC = TRUE, # setting WAIC computation to TRUE
                        nsimWAIC = 100 # number of draws for the computation of the WAIC set to 100
                        )

# summary
summary(tie_hmc)
```

--------------------------------

### Calculate AIC for remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/aic.html

This snippet shows how to calculate the AIC value for a 'remstimate' object after model estimation. Ensure the 'remstimate' object is correctly prepared and estimated.

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")
  
# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh, 
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "MLE",
                                  ncores = 1)

# AIC
aic(tie_mle) #
#> [1] 1218.625
```

--------------------------------

### Calculate Information Criteria for REM

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Use these functions to compute AIC, AICC, BIC, and WAIC for a fitted REM object. Ensure the model object (e.g., ao_mle) is already defined.

```R
aic(ao_mle)
```

```R
aicc(ao_mle)
```

```R
bic(ao_mle)
```

```R
waic(ao_mle)
```

--------------------------------

### waic Function

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/waic.html

The `waic` function computes the WAIC (Watanabe-Akaike's Information Criterion) for a given 'remstimate' object. It is an S3 method designed for objects of class 'remstimate'.

```APIDOC
## waic Function

### Description
Computes the WAIC (Watanabe-Akaike's Information Criterion) for a 'remstimate' object.

### Usage
```R
waic(object, ...)

# S3 method for class 'remstimate'
waic(object, ...)
```

### Arguments
* `object` (remstimate): An object of class 'remstimate'.
* `...`: Further arguments to be passed to the 'waic' method.

### Value
Returns the WAIC value of the 'remstimate' object.
```

--------------------------------

### Perform Diagnostics for REM

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Generate diagnostic statistics for a fitted REM object. This requires the fitted model, the event data (reh), and the statistics object (stats).

```R
ao_mle_diagnostics <- diagnostics(object = ao_mle, reh = ao_reh, stats = ao_stats)
```

--------------------------------

### plot.remstimate Function

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/plot.remstimate.html

Generates diagnostic plots for a 'remstimate' object. The type of plots generated depends on the 'approach' attribute of the object and the 'which' argument.

```APIDOC
## plot.remstimate

### Description
A function that returns a plot of diagnostics given a 'remstimate' object and depending on the 'approach' attribute.

### Usage
```R
plot(x, reh, diagnostics = NULL, which = c(1:4), effects = NULL, sender_effects = NULL, receiver_effects = NULL, ...)
```

### Arguments

*   **x** (`remstimate` object): The input remestimate object for which to generate plots.
*   **reh** (`remify` object): The remify object that was used to create the 'remstimate' object.
*   **diagnostics** (`'diagnostics' 'remstimate'` object): An object containing diagnostic information.
*   **which** (numeric vector): Specifies which diagnostic plots to generate. Possible values are:
    *   1: Q-Q plot of waiting times and density plot of rescaled waiting times.
    *   2: Standardized Schoenfeld's residuals.
    *   3: Distributions of posterior draws (histograms, for BSIR and HMC methods).
    *   4: Trace plots of posterior draws after thinning (for HMC method).
    Defaults to `c(1:4)`.
*   **effects** (character vector, optional): For tie-oriented models, the names of statistics for which to plot diagnostics. Defaults to all available statistics.
*   **sender_effects** (character vector, optional): For actor-oriented models, the names of sender statistics for which to plot diagnostics. Defaults to all available sender statistics.
*   **receiver_effects** (character vector, optional): For actor-oriented models, the names of receiver statistics for which to plot diagnostics. Defaults to all available receiver statistics.
*   **...**: Further arguments to be passed to the 'plot' method.

### Value
This function does not return a value. It generates plots as a side effect.

### Examples

```R
# Example for a tie-oriented model ('MLE')

# Load data
data(tie_data)

# Process event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")

# Specify linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# Calculate statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh, tie_effects = tie_model)

# Note: The actual plotting command using plot.remstimate would follow here,
# but requires a 'remstimate' object which is not fully constructed in this snippet.
# Example: plot(remestimate_object, tie_reh)
```
```

--------------------------------

### Calculate Statistics with remstats

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Compute the network statistics defined in the linear predictor using the `remstats` function. This function requires the processed relational event network (`reh`) and the model formula (`tie_effects`).

```R
# calculating statistics (with `remstats`)
tie_stats <- remstats::remstats(reh = tie_reh, tie_effects = tie_model)

# the 'tomstats' 'remstats' object
tie_stats
```

--------------------------------

### Generate REM Diagnostics

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Generate diagnostic information for a fitted REM object. This function requires the fitted object, the 'reh' object, and the 'stats' object.

```r
# diagnostics
tie_gd_diagnostics <- diagnostics(object = tie_gd, reh = tie_reh, stats = tie_stats)
```

```r
# diagnostics
tie_bsir_diagnostics <- diagnostics(object = tie_bsir, reh = tie_reh, stats = tie_stats)
```

--------------------------------

### Calculate WAIC for remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Compute the Widely Applicable Information Criterion (WAIC) for a fitted `remstimate` model. WAIC is a Bayesian model comparison criterion.

```R
#waic 
waic(tie_mle)
```

--------------------------------

### Calculate BIC for remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/bic.html

Use this function to retrieve the BIC value from a fitted remstimate object. This is typically done after model estimation to evaluate model fit.

```R
# BIC
bic(tie_mle)
#> [1] 1229.045
```

--------------------------------

### aic Function for remestimate Objects

Source: https://tilburgnetworkgroup.github.io/remstimate/reference/aic.html

Calculates and returns the Akaike's Information Criterion (AIC) value for a 'remstimate' object. This is useful for comparing different models fitted to the same data.

```APIDOC
## aic

A function that returns the AIC (Akaike's Information Criterion) value in a 'remstimate' object.

### Usage
```R
aic(object, ...)

# S3 method for class 'remstimate'
aic(object, ...)
```

### Arguments

* `object` (remstimate): An object of class 'remstimate'.
* `...`: Further arguments to be passed to the 'aic' method.

### Value

AIC value of a 'remstimate' object.

### Methods (by class)

* `aic(remstimate)`: AIC (Akaike's Information Criterion) value of a 'remstimate' object.

### Examples

```R
# loading data
data(tie_data)

# processing event sequence with remify
tie_reh <- remify::remify(edgelist = tie_data$edgelist, model = "tie")

# specifying linear predictor
tie_model <- ~ 1 +
               remstats::indegreeSender()+
               remstats::inertia()+
               remstats::reciprocity()

# calculating statistics
tie_reh_stats <- remstats::remstats(reh = tie_reh,
                                    tie_effects = tie_model)

# running estimation
tie_mle <- remstimate::remstimate(reh = tie_reh,
                                  stats = tie_reh_stats,
                                  method = "MLE",
                                  ncores = 1)

# AIC
aic(tie_mle) #
#> [1] 1218.625
```
```

--------------------------------

### Calculate BIC for remstimate Object

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Compute the Bayesian Information Criterion (BIC) for a fitted `remstimate` model. BIC is another measure of model fit that penalizes model complexity, with a stronger penalty than AIC.

```R
# bic 
bic(tie_mle)
```

--------------------------------

### Perform Diagnostics for GDADAMAX REM

Source: https://tilburgnetworkgroup.github.io/remstimate/articles/remstimate.html

Calculate diagnostic statistics for a REM model fitted with GDADAMAX. This requires the fitted model object, event data, and statistics object.

```R
ao_gd_diagnostics <- diagnostics(object = ao_gd, reh = ao_reh, stats = ao_stats)
# plot
# plot(x = ao_gd, reh  = ao_reh, diagnostics = ao_gd_diagnostics) # uncomment to use the plot function
```