### Example Usage of CMAverse Package

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

This example demonstrates how to load the CMAverse library and clear the environment. It is typically used for setting up an analysis session.

```R
if (FALSE) {
rm(list=ls())
library(CMAverse)
}
```

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

### Basic Usage of svymultinom

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Demonstrates how to load the CMAverse library and prepare for using the svymultinom function. This example is conditional and typically run in an interactive session.

```R
if (FALSE) {
rm(list = ls())
library(CMAverse)

}
```

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

### Install CMAverse Package

Source: https://github.com/bs1125/cmaverse/blob/master/docs/index.html

Installs the latest version of the CMAverse package from GitHub. Ensure you have the devtools package installed.

```r
devtools::install_github("BS1125/CMAverse")
```

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

### Example Usage of cmest_multistate

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmest_multistate.html

Demonstrates how to call the cmest_multistate function with specified data, parameters, and model settings. Ensure the CMAverse library is loaded before use.

```R
if (FALSE) {
library(CMAverse)
multistate_out <- cmest_multistate(data = sc_data, 
s = s_vec,
multistate_seed = 1,
exposure = 'A', mediator = 'M', outcome = 'S',
yevent = "ind_S", mevent = "ind_M",
basec = c("C1", "C2"),
basecval = c("C1" = "1", "C2" = as.character(mean(sc_data$C2))),
astar="0", a="1", 
nboot=1, EMint=F, 
bh_method = "breslow")
multistate_out
}
```

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

### Perform Sensitivity Analysis and Plot

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/ggcmsens.html

This example demonstrates how to perform a sensitivity analysis using cmest and then prepare the data for plotting with ggcmsens. It sets up a statistical model with imputation and bootstrap inference.

```R
naive <- cmest(data = cma2020, model = "rb", outcome = "contY", 
        mediator = c("M1", "M2"), 
        basec = c("C1", "C2"), EMint = TRUE,
        mreg = list("logistic", "multinomial"), yreg = "linear",
        astar = 0, a = 1, mval = list(0, "M2_0"),
        estimation = "imputation", inference = "bootstrap", nboot = 10)
```

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

### GLM for Initial Model Setup

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multiple_imputation.html

This snippet shows the initial generalized linear model call and its summary statistics, likely used for setting up the imputation process or as a baseline. It includes model call, deviance residuals, and coefficients.

```R
## Call:
glm(formula = M ~ A + C1 + C2, family = binomial(), data = getCall(x$reg.output[[3L]]$mreg[[1L]])$data, 
    weights = getCall(x$reg.output[[3L]]$mreg[[1L]])$weights)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-3.2771   0.1135   0.1611   0.2478   0.5803  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept)   0.1175     0.6424   0.183  0.85487    
A             1.7076     0.1451  11.765  < 2e-16 ***
C1            2.3329     0.6511   3.583  0.00034 ***
C2            0.9053     0.1351   6.699  2.1e-11 ***
--- 
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 2279.3  on 9999  degrees of freedom
Residual deviance: 2054.9  on 9996  degrees of freedom
AIC: 2062.9

Number of Fisher Scoring iterations: 7
```

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

### Case-Control Mediation Analysis Setup

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/case_control.html

Sets up and runs a case-control mediation analysis with a rare outcome. Specifies logistic regression for both mediator and outcome, and uses parametric function estimation with delta method inference.

```R
res_yrare <- cmest(data = data, model = "rb", casecontrol = TRUE, yrare = TRUE,
                       outcome = "Y", exposure = "A",
                       mediator = "M", basec = c("C1", "C2"), EMint = TRUE,
                       mreg = list("logistic"), yreg = "logistic",
                       astar = 0, a = 1, mval = list(1),
                       estimation = "paramfunc", inference = "delta")

summary(res_yrare)
```

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

### Weighting-based Mediation Analysis Setup

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multiple_mediators.html

This snippet demonstrates how to set up a weighting-based causal mediation analysis with multiple mediators (M1, M2) and interaction terms. It specifies logistic regression for outcome and exposure, and uses bootstrap for inference.

```R
res_wb <- cmest(data = data, model = "wb", outcome = "Y", exposure = "A",
                    mediator = c("M1", "M2"), basec = c("C1", "C2"), EMint = TRUE,
                    ereg = "logistic", yreg = "logistic",
                    astar = 0, a = 1, mval = list(0, 2), 
                    estimation = "imputation", inference = "bootstrap", nboot = 2)
```

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

### ggcmsens Function Call

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/ggcmsens.html

Example of calling the ggcmsens function with specified parameters for sensitivity analysis. Ensure the 'cmsens.html' is accessible and parameters are correctly defined.

```r
x <- [cmsens](cmsens.html)(object = naive, sens = "me", MEmethod = "rc", 
MEvariable = "C1", MEvartype = "con",
MEerror = [c](https://rdrr.io/r/base/c.html)(0.1, 0.2))
```

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

### Plot causal effect estimates with flipped coordinates

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/ggcmest.html

This example shows how to use ggcmest with coord_flip to reverse the axes, which can be useful for certain types of plots.

```r
ggcmest(x) +
coord_flip(xlim = NULL, ylim = NULL, expand = TRUE, clip = "on")
```

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

### G-Formula Model Setup

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/post_exposure_confounding.html

Sets up a causal mediation analysis using the g-formula approach with specified outcome, exposure, mediator, and covariates. It configures regression models for outcome, mediator, and post-exposure covariates, and specifies estimation and inference methods.

```R
res_gformula <- cmest(data = data, model = "gformula", outcome = "Y", exposure = "A",
                          mediator = "M", basec = c("C1", "C2"), postc = "L", EMint = TRUE,
                          mreg = list("logistic"), yreg = "logistic", postcreg = list("linear"),
                          astar = 0, a = 1, mval = list(1), 
                          estimation = "imputation", inference = "bootstrap", nboot = 2)
```

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

### Multiple Mediator Mediation Analysis (rb model, bootstrap inference)

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmest.html

Conducts mediation analysis with multiple mediators using the 'rb' model. This example uses bootstrap for inference and specifies different regression models for each mediator. Note: 10 boots are used for illustration; a higher number is recommended for actual analysis.

```R
library(CMAverse)

# multiple-mediator case with rb
# 10 boots are used for illustration
exp3 <- cmest(data = cma2020, model = "rb", outcome = "contY", 
exposure = "A", mediator = c("M1", "M2"), basec = c("C1", "C2"), 
EMint = TRUE, mreg = list("logistic", "multinomial"), 
yreg = "linear", astar = 0, a = 1, mval = list(0, "M2_0"), 
estimation = "imputation", inference = "bootstrap", nboot = 10,
boot.ci.type = "bca")
```

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

### Simulate Data with Multiple Mediator Types

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multiple_mediators.html

Simulates a dataset with a continuous confounder (C1), a binary confounder (C2), a binary exposure (A), a count mediator (M1), a categorical mediator (M2), and a binary outcome (Y). This setup is useful for testing mediation analysis with diverse variable types.

```R
set.seed(1)
expit <- function(x) exp(x)/(1+exp(x))
n <- 10000
C1 <- rnorm(n, mean = 1, sd = 0.1)
C2 <- rbinom(n, 1, 0.6)
A <- rbinom(n, 1, expit(0.2 + 0.5*C1 + 0.1*C2))
M1 <- rpois(n, exp(1 - 2*A + 0.5*C1 + 0.8*C2))
linpred1 <- 0.1 + 0.1*A + 0.4*M1 - 0.5*C1 + 0.1*C2
linpred2 <- 0.4 + 0.2*A - 0.1*M1 - C1 + 0.5*C2
probm0 <- 1 / (1 + exp(linpred1) + exp(linpred2))
probm1 <- exp(linpred1) / (1 + exp(linpred1) + exp(linpred2))
probm2 <- exp(linpred2) / (1 + exp(linpred1) + exp(linpred2))
M2 <- factor(sapply(1:n, FUN = function(x) sample(c(0, 1, 2), size = 1, replace = TRUE,
                                                 prob=c(probm0[x],
                                                        probm1[x],
                                                        probm2[x]))))
Y <- rbinom(n, 1, expit(1 + 0.8*A - 1.8*M1 + 0.5*(M2 == 1) + 0.8*(M2 == 2) + 
                              0.5*A*M1 - 0.4*A*(M2 == 1) - 1.4*A*(M2 == 2)  + 0.3*C1 - 0.6*C2))
data <- data.frame(A, M1, M2, Y, C1, C2)
```

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

### Rcreg Example: Linear Regression with Measurement Error

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/rcreg.html

Demonstrates simulating data for a linear regression model with a measurement error variable, fitting a naive model, and then applying the rcreg function for corrected analysis. Includes examples of extracting coefficients, variance-covariance matrix, residual standard deviation, formula, family, predictions, model frame, updating the model, and summary.

```R
if (FALSE) {
[rm](https://rdrr.io/r/base/rm.html)(list\]=[ls](https://rdrr.io/r/base/ls.html)())
[library](https://rdrr.io/r/base/library.html)([CMAverse](https://bs1125.github.io/CMAverse/))

# 2 boots are used for illustration
# lm
n <- 1000
x1 <- [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 5, sd = 3)
x2_true <- [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 2, sd = 1)
error1 <- [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 0, sd = 0.5)
x2_error <- x2_true + error1
x3 <- [rbinom](https://rdrr.io/r/stats/Binomial.html)(n, size = 1, prob = 0.4)
y <- 1 + 2 * x1 + 4 * x2_true + 2 * x3  + [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 0, sd = 2)
data <- [data.frame](https://rdrr.io/r/base/data.frame.html)(x1 = x1, x2_true = x2_true, x2_error = x2_error,
                   x3 = x3, y = y)
reg_naive <- [lm](https://rdrr.io/r/stats/lm.html)(y ~ x1 + x2_error + x3, data = data)
reg_true <- [lm](https://rdrr.io/r/stats/lm.html)(y ~ x1 + x2_true + x3, data = data)
reg_rc <- rcreg(reg = reg_naive, data = data, MEvariable = "x2_error", MEerror = 0.5, 
variance = TRUE, nboot = 2)
[coef](https://rdrr.io/r/stats/coef.html)(reg_rc)
[vcov](https://rdrr.io/r/stats/vcov.html)(reg_rc)
[sigma](https://rdrr.io/r/stats/sigma.html)(reg_rc)
[formula](https://rdrr.io/r/stats/formula.html)(reg_rc)
[family](https://rdrr.io/r/stats/family.html)(reg_rc)
[predict](https://rdrr.io/r/stats/predict.html)(reg_rc, newdata = data[1, ])
reg_rc_model <- [model.frame](https://rdrr.io/r/stats/model.frame.html)(reg_rc)
reg_rc_update <- [update](https://rdrr.io/r/stats/update.html)(reg_rc, data = data, weights = [rep](https://rdrr.io/r/base/rep.html)(1, n))
reg_rc_summ <- [summary](https://rdrr.io/r/base/summary.html)(reg_rc)

#glm
n <- 1000
x1 <- [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 0, sd = 1)
x2_true <- [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 1, sd = 1)
error1 <- [rnorm](https://rdrr.io/r/stats/Normal.html)(n, mean = 0, sd = 0.5)
x2_error <- x2_true + error1
x3 <- [rbinom](https://rdrr.io/r/stats/Binomial.html)(n, size = 1, prob = 0.4)
linearpred <- 1 + 0.3 * x1 - 0.5 * x2_true - 0.2 * x3
py <- [exp](https://rdrr.io/r/base/Log.html)(linearpred) / (1 + [exp](https://rdrr.io/r/base/Log.html)(linearpred))
y <- [rbinom](https://rdrr.io/r/stats/Binomial.html)(n, size = 1, prob = py)
data <- [data.frame](https://rdrr.io/r/base/data.frame.html)(x1 = x1, x2_true = x2_true, x2_error = x2_error,
                   x3 = x3, y = y)
reg_naive <- [glm](https://rdrr.io/r/stats/glm.html)(y ~ x1 + x2_error + x3, data = data, family = [binomial](https://rdrr.io/r/stats/family.html)("logit"))
reg_true <- [glm](https://rdrr.io/r/stats/glm.html)(y ~ x1 + x2_true + x3, data = data, family = [binomial](https://rdrr.io/r/stats/family.html)("logit"))
reg_rc <- rcreg(reg = reg_naive, data = data, 
MEvariable = "x2_error", MEerror = 0.5, variance = TRUE, nboot = 2)
}
```

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

### Update simexreg Model

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Use update() to modify and refit a simexreg model. This function is useful for iterating on model specifications without starting from scratch.

```R
update(simexreg)
```

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

### print.summary.svymultinom

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Prints the summary of a svymultinom object with specified precision.

```APIDOC
## print.summary.svymultinom

### Description
Prints the summary of a svymultinom object nicely with specified digits.

### Usage
```r
print(x, digits = 4, ...)
```

### Arguments
* `x` (summary.svymultinom): An object of class 'summary.svymultinom'.
* `digits` (numeric): Minimal number of significant digits. See `print.default`.
* `...`: Additional arguments.
```

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

### Summary of Initial Regression Model

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multiple_imputation.html

This output shows the summary of an initial regression model, likely before imputation, including coefficients, standard errors, and significance codes.

```R
## -3.2892   0.1105   0.1631   0.2451   0.5819  
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)    
## (Intercept)   0.1842     0.6422   0.287 0.774268    
## A             1.7317     0.1457  11.886  < 2e-16 ***
## C1            2.2255     0.6499   3.424 0.000617 ***
## C2            0.9759     0.1358   7.184 6.75e-13 ***
## --- 
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 2286.6  on 9999  degrees of freedom
## Residual deviance: 2052.2  on 9996  degrees of freedom
## AIC: 2060.2
## 
## Number of Fisher Scoring iterations: 7
```

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

### Categorical Measurement Error Analysis

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/quickstart.html

Performs a SIMEX regression for categorical variables with measurement error. This snippet shows the setup and results for a model with a single variable 'A' measured with error.

```R
simexreg(reg = getCall(x$sens[[2L]]$reg.output$mreg[[1L]])$reg, 
    formula = M2 ~ A + C1 + C2, data = getCall(x$sens[[2L]]$reg.output$mreg[[1L]])$data, 
    MEvariable = "A", MEvartype = "categorical", MEerror = c(0.95, 
    0.05, 0.05, 0.95), variance = TRUE, lambda = c(0.5, 1, 1.5, 
    2), B = 5, weights = getCall(x$sens[[2L]]$reg.output$mreg[[1L]])$weights)

## Naive coefficient estimates: 
##  (Intercept)            A           C1           C2 
##  0.003446339  0.828694328 -0.984453987  0.892199087 
## 
## Naive var-cov estimates: 
##             (Intercept)           A          C1          C2
## (Intercept)  0.30025921 -0.07438249 -0.08381839 -0.17194314
## A           -0.07438249  0.20860406 -0.00266003 -0.01458923
## C1          -0.08381839 -0.00266003  0.08577848 -0.01275232
## C2          -0.17194314 -0.01458923 -0.01275232  0.26197192
## 
## Variable measured with error:
## A
## Measurement error:
##      0    1
## 0 0.95 0.05
## 1 0.05 0.95
## 
## Error-corrected results:
##             Estimate Std. Error t value Pr(>|t|)   
## (Intercept)   0.1671     0.5520   0.303   0.7628   
## A1            0.6279     0.5216   1.204   0.2317   
## C1           -0.9362     0.2899  -3.230   0.0017 **
## C2            0.7649     0.5134   1.490   0.1396   
## --- 
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
```

```R
simexreg(reg = getCall(x$sens[[2L]]$reg.output$mreg[[2L]])$reg, 
    formula = M2 ~ A + C1 + C2, data = getCall(x$sens[[2L]]$reg.output$mreg[[2L]])$data, 
    MEvariable = "A", MEvartype = "categorical", MEerror = c(0.9, 
    0.1, 0.1, 0.9), variance = TRUE, lambda = c(0.5, 1, 1.5, 
    2), B = 5, weights = getCall(x$sens[[2L]]$reg.output$mreg[[2L]])$weights)

## Naive coefficient estimates: 
## (Intercept)           A          C1          C2 
##   1.6402566   0.2901364   0.6208107   2.0833115 
## 
## Naive var-cov estimates: 
##             (Intercept)            A            C1            C2
## (Intercept)  0.06573913 -0.018924346 -0.0173562931 -0.0381103530
## A           -0.01892435  0.047073002  0.0032931500 -0.0062472874
## C1          -0.01735629  0.003293150  0.0141472682  0.0003964488
## C2          -0.03811035 -0.006247287  0.0003964488  0.0559647177
## 
## Variable measured with error:
## A
## Measurement error:
##     0   1
## 0 0.9 0.1
## 1 0.1 0.9
## 
## Error-corrected results:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   1.5985     0.2803   5.702 1.30e-07 ***
## A             0.3403     0.3206   1.062    0.291    
## C1            0.6341     0.1211   5.235 9.71e-07 ***
## C2            2.0890     0.2390   8.741 7.49e-14 ***
## --- 
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
```

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

### Summary Method for cmsens.me Objects

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmsens.html

S3 method for summarizing sensitivity analysis results related to measurement error.

```R
summary(object, ...)
```

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

### Summarize simexreg Results Nicely

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Use summary() to obtain a detailed summary of the simexreg object. This includes coefficient estimates, standard errors, p-values, and other diagnostic statistics.

```R
summary(simexreg)
```

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

### Causal Mediation Analysis with Error Correction

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/measurement_error.html

Performs causal mediation analysis using the `cmest` function. This is useful for estimating direct and indirect effects when there might be measurement error. Ensure the `cmest` package is installed and loaded.

```R
res_true <- cmest(data = data, model = "rb", outcome = "Y", exposure = "A",
                           mediator = "M", basec = c("C1", "C2"), EMint = TRUE,
                           mreg = list("logistic"), yreg = "logistic",
                           astar = 0, a = 1, mval = list(1), 
                           estimation = "paramfunc", inference = "delta")
```

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

### Prepare Data for svymultinom

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Generates synthetic data with independent variables (x1, x2, x3), probabilities for three categories (py1, py2, py3), a categorical outcome variable (y), and survey weights (w). This data is then formatted into a data frame for use with svymultinom.

```R
n <- 1000
x1 <- rnorm(n, mean = 0, sd = 1)
x2 <- rnorm(n, mean = 1, sd = 1)
x3 <- rbinom(n, size = 1, prob = 0.4)
linearpred1 <- 1 + 0.3 * x1 - 0.5 * x2 - 0.2 * x3
linearpred2 <- 2 + 1 * x1 - 2 * x2 - 1 * x3
py2 <- exp(linearpred1) / (1 + exp(linearpred1) + exp(linearpred2))
py3 <- exp(linearpred2) / (1 + exp(linearpred1) + exp(linearpred2))
py1 <- 1 - py2 - py3
y <- sapply(1:n, function(x) sample(size = 1, c(1:3), prob = c(py1[x], py2[x], py3[x])))
w <- ifelse(x3 == 0, 0.4, 0.6)
data <- data.frame(x1 = x1, x2 = x2, x3 = x3, y = y)
```

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

### Print Summary of simexreg Nicely

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Use print(summary.simexreg) to display the summary of a simexreg object in a nicely formatted way. This provides a concise overview of the model's performance and results.

```R
print(summary.simexreg)
```

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

### Simulate and Prepare Data for Multistate Model

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multistate_model.html

This R code simulates data and prepares it for a multistate model. It involves generating event times, merging data, and selecting relevant covariates. Ensure 'gen_srv' function and necessary variables (c1, c2, c3, c4, data_23, data) are defined.

```R
for(i in 1:dim(data_23)[1]){
      data_23_tem$id[i] = data_23$id[i]
      repeat {
        time_test = gen_srv(n = 1, 
                            lambda = 0.8, 
                            beta = c(as.numeric(c1),
                                     c2,
                                     as.numeric(c3),
                                     as.numeric(c4)), 
                            X = data_23[i, c("A.3", "M.3", "C1.3","C2.3")])
        # exit if the condition is met
        if (time_test > data_23[i,"M.3"]) break
      }
      data_23_tem$new_y[i] = time_test
    }
    data_temp = merge(data, data_23_tem, by = "id", all = T)
    # modify M and S
    data_temp$S[which(data_temp$ind_M == 1)] = data_temp$new_y[which(data_temp$ind_M == 1)]
    data_temp$M[which(data_temp$ind_M == 0)] = data_temp$S[which(data_temp$ind_M == 0)]
    data_final = data_temp
    data_final$A = as.factor(data_final$A) 
    
    sc_data = data_final %>% dplyr::select(id,A,M,S,ind_M,ind_S,C1,C2)
    
    # generate time to censoring and update event indicator
    time_to_censor = runif(n, 0, 2*max(sc_data$S))
    sc_data$ind_S = ifelse(sc_data$S > time_to_censor, 0, 1)
    sc_data$A = factor(sc_data$A)
    sc_data$C1 = factor(sc_data$C1)
```

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

### cmdag()

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/index.html

Generates a Directed Acyclic Graph (DAG) visualization.

```APIDOC
## cmdag()

### Description
DAG Visualization

### Usage
cmdag()

### Details
This function is used for visualizing Directed Acyclic Graphs, likely in the context of causal inference.

### Value
A visualization object.
```

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

### Direct Counterfactual Imputation Estimation with Bootstrap Inference

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/single_mediator.html

Use this snippet to estimate direct counterfactuals with imputation and bootstrap inference. Ensure 'cmest' is installed and data is properly formatted. Set 'nboot' to a sufficiently large number for reliable inference.

```r
res_gformula <- cmest(data = data, model = "gformula", outcome = "Y", exposure = "A",
                          mediator = "M", basec = c("C1", "C2"), EMint = TRUE,
                          mreg = list("logistic"), yreg = "logistic",
                          astar = 0, a = 1, mval = list(1), 
                          estimation = "imputation", inference = "bootstrap", nboot = 2)

summary(res_gformula)
```

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

### print.simexreg

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Prints a summary of a simexreg object.

```APIDOC
## print.simexreg

### Description
Prints a summary of a simexreg object.

### Method
`print` (S3 method for simexreg)

### Parameters
#### Arguments
- **x** (simexreg) - An object of class `simexreg`.
- **...** - Additional arguments.
```

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

### print.summary.simexreg

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Prints a summary of the summary of a simexreg object.

```APIDOC
## print.summary.simexreg

### Description
Prints a summary of the summary of a simexreg object.

### Method
`print` (S3 method for summary.simexreg)

### Parameters
#### Arguments
- **x** (summary.simexreg) - An object of class `summary.simexreg`.
- **digits** (integer) - Minimal number of significant digits. See `print.default`.
- **...** - Additional arguments.
```

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

### Simulate Data with Post-exposure Confounding

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/post_exposure_confounding.html

Simulates a dataset for mediation analysis with a continuous baseline confounder, a binary baseline confounder, a binary exposure, a continuous mediator-outcome confounder affected by the exposure, a binary mediator, and a binary outcome. This setup is crucial for testing methods that account for exposure-dependent confounders.

```R
set.seed(1)
expit <- function(x) exp(x)/(1+exp(x))
n <- 10000
C1 <- rnorm(n, mean = 1, sd = 0.1)
C2 <- rbinom(n, 1, 0.6)
A <- rbinom(n, 1, expit(0.2 + 0.5*C1 + 0.1*C2))
L <- rnorm(n, mean = 1 + A - C1 - 0.5*C2, sd = 0.5)
M <- rbinom(n, 1, expit(1 + 2*A - L + 1.5*C1 + 0.8*C2))
Y <- rbinom(n, 1, expit(-3 - 0.4*A - 1.2*M + 0.5*A*M - 0.5*L + 0.3*C1 - 0.6*C2))
data <- data.frame(A, M, Y, C1, C2, L)
```

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

### Single-Mediator Analysis (Continuous Outcome)

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmest_rb.html

Example of using cmest_rb for a single-mediator case with a continuous outcome, without exposure-mediator interaction. Requires specifying outcome, exposure, mediator, base covariates, regression models for outcome and mediator, estimation method, inference method, and specific values for exposure and mediator.

```R
if (FALSE) {
library(CMAverse)

# single-mediator case without exposure-mediator interaction
exp1 <- cmest_rb(data = cma2020, outcome = "contY", 
exposure = "A", mediator = "M1", basec = c("C1", "C2"), 
EMint = FALSE, yreg = "linear", mreg = list("logistic"), 
estimation = "paramfunc", inference = "delta", astar = 0, a = 1, mval = list(1))
summary(exp1)

}
```

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

### Extract Model Information

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Demonstrates how to extract coefficients, variance-covariance matrix, formula, and model frame from a fitted svymultinom object.

```R
coef(reg)
vcov(reg)
formula(reg)
model.frame(reg)
```

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

### Simulate Data for Causal Analysis

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/quickstart.html

Simulates a dataset with a binary exposure, two mediators (binary and continuous), a continuous outcome, and two baseline confounders. This data is used for demonstrating the package's functionalities.

```r
set.seed(1)
n <- 100
C1 <- rnorm(n, mean = 1, sd = 1)
C2 <- rbinom(n, 1, 0.6)
pa <- exp(0.2 - 0.5*C1 + 0.1*C2)/(1 + exp(0.2 - 0.5*C1 + 0.1*C2))
A <- rbinom(n, 1, pa)
pm <- exp(1 + 0.5*A - 1.5*C1 + 0.5*C2)/ (1 + exp(1 + 0.5*A - 1.5*C1 + 0.5*C2))
M1 <- rbinom(n, 1, pm)
M2 <- rnorm(n, 2 + 0.8*A - M1 + 0.5*C1 + 2*C2, 1)
Y <- rnorm(n, mean = 0.5 + 0.4*A + 0.5*M1 + 0.6*M2 + 0.3*A*M1 + 0.2*A*M2 - 0.3*C1 + 2*C2, sd = 1)
data <- data.frame(A, M1, M2, Y, C1, C2)
```

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

### Multiple Mediator Mediation Analysis (ne model, bootstrap inference)

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmest.html

Performs mediation analysis with multiple mediators using the 'ne' (natural effect) model. This example uses bootstrap for inference and explicitly defines the outcome regression model using a glm formula. Requires specifying outcome, exposure, mediators, base covariates, reference values for exposure and mediator, and estimation/inference methods.

```R
library(CMAverse)

# multiple-mediator case with ne
exp4 <- cmest(data = cma2020, model = "ne", outcome = "contY", EMint = TRUE,
exposure = "A", mediator = c("M1", "M2"), basec = c("C1", "C2"), 
yreg = glm(contY ~ A + M1 + M2 + A*M1 + A*M2 + C1 + C2, family = gaussian, data = cma2020), 
astar = 0, a = 1, mval = list(0, "M2_0"), estimation = "imputation", 
inference = "bootstrap", nboot = 10)
```

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

### Summarize svymultinom Model

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Provides a summary of the fitted svymultinom model, including coefficients, standard errors, and significance tests.

```R
summary(reg)
```

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

### summary.svymultinom

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Generates a summary of the svymultinom object, providing a concise overview of the model results.

```APIDOC
## summary.svymultinom

### Description
Summarizes the results of a svymultinom object nicely.

### Usage
```r
summary(object, ...)
```

### Arguments
* `object` (svymultinom): An object of class 'svymultinom'.
* `...`: Additional arguments.

### Value
A summary object that can be printed using `print.summary.svymultinom`.
```

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

### Print Method for Summary of cmsens.me Objects

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmsens.html

S3 method for printing the summary of measurement error sensitivity analysis results.

```R
print(x, digits = 4, ...)
```

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

### Mediator Regression Analysis (Initial Dataset)

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multiple_imputation.html

This snippet shows the mediator regression model for the initial dataset before imputation. It provides a baseline for comparison.

```R
## Mediator regressions: 

## Call: 
glm(formula = M ~ A + C1 + C2, family = binomial(), data = getCall(x$reg.output[[1L]]$mreg[[1L]])$data, 
    weights = getCall(x$reg.output[[1L]]$mreg[[1L]])$weights)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-3.2820   0.1137   0.1633   0.2434   0.5946  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept) -0.03518    0.64316  -0.055 0.956376    
A            1.66962    0.14384  11.607  < 2e-16 ***
C1           2.48028    0.65225   3.803 0.000143 ***
C2           0.94592    0.13553   6.980 2.96e-12 ***
--- 
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 2030.4  on 9999  degrees of freedom
Residual deviance: 1973.2  on 9994  degrees of freedom
AIC: 1985.2

Number of Fisher Scoring iterations: 7
```

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

### print.svymultinom

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Prints the results of a svymultinom object in a user-friendly format.

```APIDOC
## print.svymultinom

### Description
Prints the results of a svymultinom object nicely.

### Usage
```r
print(x, ...)
```

### Arguments
* `x` (svymultinom): An object of class 'svymultinom'.
* `...`: Additional arguments.

### Methods
* `print(summary.svymultinom)`: Print summary of `svymultinom` nicely.
```

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

### Extract Information from simexreg Object (Linear Model)

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Demonstrates how to extract coefficients, variance-covariance matrix, standard deviation, formula, family, predictions, model frame, updated model, and summary from a simexreg object fitted to a linear model.

```R
coef(reg_simex)
vcov(reg_simex)
sigma(reg_simex)
formula(reg_simex)
family(reg_simex)
predict(reg_simex, newdata = data[1, ])
reg_simex_model <- model.frame(reg_simex)
reg_simex_update <- update(reg_simex, data = data, weights = rep(1, n))
reg_simex_summ <- summary(reg_simex)
```

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

### Print simexreg Results Nicely

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/simexreg.html

Use print() to display the results of a simexreg object in a nicely formatted way. This function provides a user-friendly output of the model's key information.

```R
print(simexreg)
```

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

### Predict Probabilities with svymultinom

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/svymultinom.html

Generates predicted probabilities for the first observation in the dataset using the fitted svymultinom model.

```R
predict(reg, newdata = data[1, ], type = "probs")
```

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

### Accessing Bootstrap Summary

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multistate_model.html

Retrieves the summary of bootstrapped RD, SD, and TE estimates, including percentiles, from the cmest_multistate function output.

```r
sc_data_result$bootstrap_summary
```

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

### Data Simulation for Multistate Model

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/multistate_model.html

Simulates a dataset for a multistate model, including a binary exposure, time-to-event mediator and outcome, and baseline covariates. Assumes exponential time-to-event distributions.

```R
# set up coefficients 
# transition 1 (A to M)
a1 = -1.9 
a2 = 0.2 
a3 = 0.5 

# transition 2 (A to S)
b1 = 1 
b3 = -0.5
b4 = 0.3

# transition 3 (M to S)
c1 = 0.55
c2 = -0.15 
c3 = -0.1 
c4 = -0.2 

set.seed(8) 

# build a function to generate time-to-event data
gen_srv <- function(n, lambda, beta, X){
  X = as.matrix(X)
  beta = as.matrix(beta, ncol=1)
  time = -log(runif(n)) / (lambda * exp(X %*% beta)) 
  return(time) 
}

n <- 2000 
A = sample(c(0,1),replace=TRUE, size=n, c(0.5,0.5)) #binary exposure
C1 = sample(c(0,1),replace=TRUE, size=n,c(0.6, 0.4)) #binary baseline covariate
C2 = rnorm(n, mean = 1, sd = 1) #continuous baseline covariate
id=c(1:n)
full = data.frame(id,A,C1,C2)
M = gen_srv(n=n, lambda = 1.5, beta = c(a1,a2,a3), X=data.frame(A,C1,C2)) #time to event mediator
S = gen_srv(n=n, lambda = 1, beta = c(b1,b3,b4), X=data.frame(A,C1,C2)) #time to event outcome
data = data.frame(id = c(1:n), M = M, S = S)
# indicator for event
data$ind_M = ifelse(data$M <= data$S, 1, 0)
data$ind_S = 1
data <- merge(data,full , by = "id")
# modify S distribution
trans_matrix = transMat(x = list(c(2, 3), c(3), c()), names = c("A", "M", "S"))
covs = c("A","M", "C1","C2")
pre_data = msprep(time = c(NA, "M", "S"), status = c(NA, "ind_M", "ind_S"),
                         data = data, trans = trans_matrix, keep = covs)
pre_data = expand.covs(pre_data, covs, append = TRUE, longnames = FALSE)
# resample for T < S
data_23 = pre_data[which(pre_data$trans == 3),]
data_23_tem = data.frame(id = rep(NA,dim(data_23)[1]),
                         new_y = rep(NA,dim(data_23)[1]))
paste("# to resample is ", nrow(data_23))
```

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

### Run SIMEX for Continuous Measurement Error

Source: https://github.com/bs1125/cmaverse/blob/master/docs/articles/measurement_error.html

Initiates a sensitivity analysis for a continuous variable measured with error using the SIMEX method. Requires a pre-existing naive model result (`res_naive_cont`), the name of the variable with error (`MEvariable`), its type (`MEvartype`), and the estimated measurement error (`MEerror`).

```R
res_simex_cont <- cmsens(object = res_naive_cont, sens = "me", MEmethod = "simex", 
                             MEvariable = "C1_error", MEvartype = "con", MEerror = 0.05)
```

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

### Plot DAG with cmdag (With Base and Post-Exposure Confounders)

Source: https://github.com/bs1125/cmaverse/blob/master/docs/reference/cmdag.html

Use this snippet to visualize a DAG that includes both base and post-exposure confounders. Node and text colors are customizable.

```R
cmdag(outcome = "Y", exposure = "A", mediator = c("M1", "M2"),
     basec = c("C1", "C2", "C3"), postc = c("L1", "L2"), node = FALSE, text_col = "black")
```