### Install development version from GitHub

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/index.html

Installation method for the latest development version using the devtools package.

```R
# install.packages("devtools")
devtools::install_github("smartdata-analysis-and-statistics/metamisc")
```

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

### Install metamisc from CRAN

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/index.html

Standard installation method for the stable version of the package from the CRAN repository.

```R
install.packages("metamisc")
```

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

### Build Generalizable Models

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/metapred.html

Examples for building generalizable models using the metapred function with various parameters.

```R
metapred(DVTipd, strata = "study", formula = dvt ~ 1, scope = f, family = binomial) # Forwards
metapred(DVTipd, strata = "study", formula = f, scope = f, family = binomial) # no selection
metapred(DVTipd, strata = "study", formula = f, max.steps = 0, family = binomial) # no selection
metapred(DVTipd, strata = "study", formula = f, recal.int = TRUE, family = binomial)
metapred(DVTipd, strata = "study", formula = f, meta.method = "REML", family = binomial)
```

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

### Generating Forest Plots for Performance Estimates

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.mm_perf.html

Examples demonstrating how to calculate performance metrics and visualize them using forest plots.

```R
data(EuroSCORE)

# Calculate the c-statistic and its standard error
est1 <- ccalc(cstat = c.index, cstat.se = se.c.index, cstat.cilb = c.index.95CIl, 
              cstat.ciub = c.index.95CIu, N = n, O = n.events, data = EuroSCORE, slab = Study)
plot(est1)


# Calculate the total O:E ratio and its standard error
est2 <- oecalc(O = n.events, E = e.events, N = n, data = EuroSCORE, slab = Study)
plot(est2)
```

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

### Recalibrating a metapred model

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/recalibrate.html

Example demonstrating how to recalibrate a metapred model using a subset of data.

```R
data(DVTipd)
DVTipd$cluster <- 1:4 # Add a fictional clustering to the data set.
# Suppose we estimated the model in three studies: 
DVTipd123 <- DVTipd[DVTipd$cluster <= 3, ]
mp <- metamisc:::metapred(DVTipd123, strata = "cluster", f = dvt ~ vein + malign, 
family = binomial)
# and now want to recalibrate it for the fourth:
DVTipd4 <- DVTipd[DVTipd$cluster == 4, ]
metamisc:::recalibrate(mp, newdata = DVTipd4)
```

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

### Display meta-analysis summary output

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pm.html

Example output showing the summary c-statistic with confidence and prediction intervals.

```text
Summary c-statistic with 95% confidence and (approximate) 95% prediction interval:

 Estimate       CIl       CIu       PIl       PIu
0.7888603 0.7648784 0.8110005 0.6792568 0.8682736

Number of studies included:  23
```

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

### Generate funnel plots

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.fat.html

Examples demonstrating the creation of basic funnel plots and plots using alternative test methods.

```R
data(Fibrinogen)
b <- log(Fibrinogen$HR)
b.se <- ((log(Fibrinogen$HR.975) - log(Fibrinogen$HR.025))/(2*qnorm(0.975)))
n.total <- Fibrinogen$N.total

# A very simple funnel plot
plot(fat(b=b, b.se=b.se), xlab = "Log hazard ratio")


# Plot the funnel for an alternative test
plot(fat(b=b, b.se=b.se, n.total=n.total, method="M-FIV"), xlab = "Log hazard ratio")
```

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

### Generate meta-analysis distribution plots

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/dplot.uvmeta.html

Examples demonstrating how to visualize prior and posterior distributions for model parameters using the dplot function.

```R
if (FALSE) { # \dontrun{
data(Roberts)

fit <- with(Roberts, uvmeta(r=SDM, r.se=SE, method="BAYES"))

dplot(fit)
dplot(fit, distr_type = "posterior")
dplot(fit, par = "tau", distr_type = "prior")
dplot(fit, plot_type = "hist")
} # }
```

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

### Example: Plotting a Frequentist Meta-analysis

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.uvmeta.html

Demonstrates creating a forest plot from a frequentist random-effects meta-analysis using the Roberts dataset.

```R
data(Roberts)

# Frequentist random-effects meta-analysis
fit <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts)))
plot(fit) 
```

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

### Generating a forest plot for a metapred fit

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/forest.metapred.html

Example demonstrating the creation of a metapred fit and the subsequent attempt to generate a forest plot.

```R
data(DVTipd)

# Internal-external cross-validation of a pre-specified model 'f'
f <- dvt ~ histdvt + ddimdich + sex + notraum
fit <- metapred(DVTipd, strata = "study", formula = f, scope = f, family = binomial)

# Display the model's external performance (expressed as mean squared error by default) 
# for each study
forest(fit)
#> Error in forest.default(fit): Must specify either 'vi', 'sei', or ('ci.lb','ci.ub').
```

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

### Example: Plotting autocorrelation for a fitted model

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/acplot.valmeta.html

Demonstrates fitting a Bayesian meta-analysis model using valmeta and visualizing the autocorrelation of the results.

```R
if (FALSE) { # \dontrun{
data(EuroSCORE)

fit <- valmeta(cstat=c.index, cstat.se=se.c.index, cstat.cilb=c.index.95CIl,
               cstat.ciub=c.index.95CIu, N=n, O=n.events, 
               data=EuroSCORE, method="BAYES", slab=Study)
acplot(fit)
} # }
```

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

### Example: Imputing missing values

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/impute_conditional_mean.html

Demonstrates defining population parameters and imputing missing values in a partially observed sample.

```R
# Define the population means
mu <- c(0, 1, 2)

# Define the covariance of the population
Sigma <- diag(1,3)
Sigma[1,2] <- Sigma[2,1] <- 0.3 
Sigma[2,3] <- Sigma[3,2] <- 0.1
Sigma[1,3] <- Sigma[3,1] <- -0.2

# Generate a 'random' sample from the population that is partially observed
x <- c(NA, 2, 4)

# Impute the missing values
impute_conditional_mean (x=x, mu=mu, Sigma=Sigma)
#> [1] -0.1414141  2.0000000  4.0000000
```

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

### Example: Plotting running means from a Bayesian meta-analysis

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/rmplot.uvmeta.html

Demonstrates fitting a Bayesian meta-analysis model using the Roberts dataset and visualizing the results with rmplot.

```R
if (FALSE) { # \dontrun{
data(Roberts)

fit <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts), method="BAYES"))
rmplot(fit)
} # }
```

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

### Example usage of gelmanplot

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/gelmanplot.valmeta.html

Demonstrates how to generate a plot after performing a Bayesian meta-analysis of the concordance statistic.

```R
if (FALSE) { # \dontrun{
data(EuroSCORE)

# Meta-analysis of the concordance statistic
fit <- valmeta(cstat=c.index, cstat.se=se.c.index, cstat.cilb=c.index.95CIl,
               cstat.ciub=c.index.95CIu, N=n, O=n.events, 
               data=EuroSCORE, method="BAYES", slab=Study)
gelmanplot(fit)
} # } 
```

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

### Display Bayesian meta-analysis summary output

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pm.html

Example output showing the summary c-statistic with credibility and prediction intervals from the Bayesian model.

```text
Summary c-statistic with 95% credibility and  95% prediction interval:

 Estimate       CIl       CIu       PIl       PIu
0.7885098 0.7648367 0.8123593 0.6800665 0.8775017
```

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

### Example: Plotting autocorrelation for a Bayesian meta-analysis

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/acplot.uvmeta.html

Demonstrates fitting a Bayesian meta-analysis model using the Roberts dataset and visualizing the autocorrelation of the results.

```R
if (FALSE) { # \dontrun{
data(Roberts)

fit <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts), method="BAYES"))
acplot(fit)
} # }
```

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

### Load metafor package

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/fat.html

Initializes the metafor package to access statistical testing functions.

```R
require(metafor)
```

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

### Calculate O:E ratio for binary outcome prediction models

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/oecalc.html

Example usage of the oecalc function with the EuroSCORE dataset.

```R
######### Validation of prediction models with a binary outcome #########
data(EuroSCORE)
```

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

### Extracting Models from Metapred Objects

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/subset.metapred.html

Demonstrates how to initialize a metapred object and use the subset function to extract various model configurations.

```R
data(DVTipd)
DVTipd$cluster <- letters[1:4] # Add a fictional clustering to the data.
mp <- metapred(DVTipd, strata = "cluster", formula = dvt ~ histdvt + ddimdich, family = binomial)
subset(mp) # best cross-validated model
```

```R
subset(mp, select = "global") # Final model fitted on all strata.
```

```R
subset(mp, step = 1) # The best model of step 1
```

```R
subset(mp, step = 1, model = "histdvt") # The model in which histdvt was removed, in step 1.
```

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

### Load DVTipd dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/DVTipd.html

Load the hypothetical dataset into the R environment.

```R
data(DVTipd)
```

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

### Load and inspect DVTipd dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/DVTipd.html

Load the dataset into the R environment and display its structure and summary statistics.

```R
data(DVTipd)
str(DVTipd) 
#> 'data.frame':	500 obs. of  16 variables:
#>  $ sex     : num  0 1 0 1 0 0 1 0 1 0 ...
#>  $ malign  : num  0 0 0 0 0 0 0 0 0 0 ...
#>  $ par     : num  0 0 1 0 0 0 0 0 0 0 ...
#>  $ surg    : num  0 0 0 0 0 0 0 0 1 0 ...
#>  $ tend    : num  1 1 0 1 1 0 0 1 1 1 ...
#>  $ oachst  : num  0 0 0 0 0 0 0 0 0 0 ...
#>  $ leg     : num  1 0 0 0 0 1 1 0 0 0 ...
#>  $ notraum : num  1 1 1 1 1 0 0 1 0 1 ...
#>  $ calfdif3: num  0 0 0 0 0 0 0 0 0 0 ...
#>  $ pit     : num  0 0 0 0 0 1 0 1 1 1 ...
#>  $ vein    : num  0 0 0 0 1 0 0 0 0 1 ...
#>  $ altdiagn: num  1 0 1 1 1 0 1 1 1 1 ...
#>  $ histdvt : num  0 1 0 0 0 0 1 0 0 0 ...
#>  $ ddimdich: num  1 0 0 0 0 1 1 0 1 1 ...
#>  $ dvt     : num  0 0 0 0 0 0 0 0 0 0 ...
#>  $ study   : Factor w/ 4 levels "a","b","c","d": 1 4 1 4 1 4 4 4 4 2 ...
summary(apply(DVTipd,2,as.factor))
#>      sex               malign              par                surg          
#>  Length:500         Length:500         Length:500         Length:500        
#>  Class :character   Class :character   Class :character   Class :character  
#>  Mode  :character   Mode  :character   Mode  :character   Mode  :character  
#>      tend              oachst              leg              notraum         
#>  Length:500         Length:500         Length:500         Length:500        
#>  Class :character   Class :character   Class :character   Class :character  
#>  Mode  :character   Mode  :character   Mode  :character   Mode  :character  
#>    calfdif3             pit                vein             altdiagn        
#>  Length:500         Length:500         Length:500         Length:500        
#>  Class :character   Class :character   Class :character   Class :character  
#>  Mode  :character   Mode  :character   Mode  :character   Mode  :character  
#>    histdvt            ddimdich             dvt               study          
#>  Length:500         Length:500         Length:500         Length:500        
#>  Class :character   Class :character   Class :character   Class :character  
#>  Mode  :character   Mode  :character   Mode  :character   Mode  :character  
```

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

### Perform univariate meta-analysis with uvmeta

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/uvmeta-class.html

Demonstrates creating a uvmeta object using effect sizes and variances from the Collins dataset.

```R
data(Collins)

#Extract effect size and error variance
r <- Collins$logOR
vars <- Collins$SE**2

#Frequentist random-effects meta-analysis
fit1 <- uvmeta(r,vars)

#Extract results
fit1$results
#> NULL
```

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

### Load Zhang Dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Zhang.html

Load the dataset into the R environment.

```R
data("Zhang")
```

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

### Load required R packages

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pf.html

Load the metafor and metamisc packages to access functions for meta-analysis of prognosis studies.

```R
library(metafor)
library(metamisc)
```

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

### plot(x, ...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.valmeta.html

Creates a forest plot for objects of class 'valmeta'.

```APIDOC
## plot(x, ...)

### Description
Generates a forest plot showing performance estimates of each validation with corresponding confidence intervals. A summary estimate polygon is included, and a 95% prediction interval is added for random-effects models.

### Arguments
- **x** (valmeta) - Required - An object of class "valmeta".
- **...** (dots) - Optional - Additional arguments passed to forest.

### Value
Returns an object of class ggplot.

### Examples
```r
data(EuroSCORE)
fit <- valmeta(cstat=c.index, cstat.se=se.c.index, cstat.cilb=c.index.95CIl,
               cstat.ciub=c.index.95CIu, N=n, O=n.events, data=EuroSCORE)
plot(fit)

library(ggplot2)
plot(fit, theme=theme_grey())
```
```

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

### View the Zhang dataset structure

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pf.html

A tabular representation of the Zhang dataset showing study metadata and statistical estimates.

```text
                   Study    PrimaryAuthor year     Country Disease   N    HR
1  Gonzalez-Rodilla 2013 Gonzalez-Rodilla 2013       Spain      EC 126  1.14
2            Togami 2012           Togami 2012       Japan    UPSC  71  3.00
3              Mori 2010             Mori 2010       Japan     EEC  63  2.58
4          Jongen 2009-2           Jongen 2009 Netherlands     EEC 315  0.66
5           Konecny 2009          Konecny 2009         USA      EC 273  2.13
6           Odicino 2008          Odicino 2008       Italy    UPSC  10  1.60
7          Santin 2005-2           Santin 2005         USA    UPSC  30  5.19
8          Santin 2005-1           Santin 2005         USA    UPSC  27 12.43
9             Backe 1997            Backe 1997     Germany      EC 222  1.17
10       Kohlberger 1996       Kohlberger 1996   Australia      EC 100  3.95
11          Saffari 2995          Saffari 1995         USA      EC  75  5.25
12            Peiro 2004            Peiro 2004     Germany      EC  10  1.64
13       Cianciulli 2003       Cianciulli 2003       Italy      EC  73  1.90
14            Gates 2006            Gates 2006         USA      EC  99  4.00
15           Togami 2012           Togami 2012       Japan    UPSC  71  3.43
16             Mori 2010             Mori 2010       Japan     EEC  63  3.80
17         Jongen 2009-2           Jongen 2009 Netherlands     EEC 315  1.75
18         Coronado 2001         Coronado 2001       Spain      EC 114  2.69
19            Backe 1997            Backe 1997     Germany      EC 222  1.36
20             Voss 2011              Vos 2011     England      EC 156  1.41
   HR.025 HR.975 outcome
1    0.52   2.52      OS
2    1.15   7.01      OS
3    1.03   7.92      OS
4    0.10   4.28      OS
5    1.65   2.73      OS
6    0.89   2.89      OS
7    1.52  17.64      OS
8    1.57  98.28      OS
9    0.93   1.49      OS
10   1.07  14.69      OS
11   1.33  20.74      OS
12   1.04   2.57      OS
13   0.83   4.36      OS
14   0.77  20.80      OS
15   1.50   7.23     PFS
16   1.29  12.35     PFS
17   0.23  13.08     PFS
18   1.35   5.37     PFS
19   0.65   2.85     PFS
20   0.51   3.88     PFS
```

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

### Alternative Meta-Analysis Syntax

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pf.html

Demonstrates the default rma function call which yields the same results as the explicit REML method.

```R
# The following approach yields the same results
# rma(yi=logHR, sei=se.logHR, subset=(outcome=="OS"), data=Zhang)
```

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

### plot(x, sort = "asc", ...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.uvmeta.html

Generates a forest plot for a 'uvmeta' object, displaying performance estimates, confidence intervals, and summary estimates.

```APIDOC
## plot(x, sort = "asc", ...)

### Description
Creates a forest plot for objects of class "uvmeta". The plot displays performance estimates of each validation with corresponding confidence intervals and a summary estimate polygon.

### Arguments
- **x** (uvmeta) - Required - An object of class "uvmeta".
- **sort** (string) - Optional - Determines study ordering. Use "asc" for ascending effect size, "desc" for descending, or any other value to ignore ordering. Defaults to "asc".
- **...** (dots) - Optional - Additional arguments passed to forest.

### Example
```R
data(Roberts)
fit <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts)))
plot(fit)
```
```

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

### Usage of summary method for riley objects

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/summary.riley.html

Standard S3 method signature for generating parameter summaries from a riley object.

```R
# S3 method for class 'riley'
summary(object, ...)
```

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

### Load Roberts dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Roberts.html

Load the Roberts dataset into the R environment for analysis.

```R
data("Roberts")
```

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

### Load Kertai dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Kertai.html

Load the Kertai dataset into the R environment for analysis.

```R
data("Kertai")
```

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

### Perform Meta-Analysis with Riley

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/riley.html

Demonstrates fitting meta-analysis models to various datasets including Daniels, Kertai, and Scheidler.

```R
data(Scheidler)
data(Daniels)
data(Kertai)

# Meta-analysis of potential surrogate markers data
# The results obtained by Riley (2008) were as follows:
# beta1 = -0.042 (SE = 0.063),
# beta2 = 14.072 (SE = 4.871)
# rho   = -0.759
if (FALSE) { # \dontrun{
fit1 <- riley(Daniels) #maxit reached, try again with more iterations
} # }
fit1 <- riley(Daniels, control=list(maxit=10000))
summary(fit1)
#> Call:
#> riley.default(X = Daniels, control = list(maxit = 10000))
#> 
#>           Estimate         SE       2.5 %     97.5 %
#> beta1  0.005298983 0.06479973 -0.12170616  0.1323041
#> beta2 13.505678310 4.99256719  3.72042644 23.2909302
#> psi1   0.134785102 0.09190903 -0.04535329  0.3149235
#> psi2  18.076027226 4.00992798 10.21671280 25.9353417
#> rho   -0.748689375 0.15266774 -0.89430728 -0.4596724

# Meta-analysis of prognostic test studies
fit2 <- riley(Kertai)
fit2
#> Call:
#> riley.default(X = Kertai)
#> 
#> Coefficients
#>      beta1      beta2       psi1       psi2        rho 
#>  0.8164679 -0.9715821  0.3499043  0.7692122  0.1537878 
#> 
#> Degrees of Freedom: 9 Residual

# Meta-analysis of computed tomography data 
ds <- Scheidler[which(Scheidler$modality==1),]
fit3 <- riley(ds)
fit3
#> Call:
#> riley.default(X = ds)
#> 
#> Coefficients
#>       beta1       beta2        psi1        psi2         rho 
#> -0.01731291 -2.32166611  0.71181410  0.38103153  0.70119871 
#> 
#> Degrees of Freedom: 29 Residual
```

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

### Perform two-stage meta-analysis of O:E ratios

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/valmeta.html

Conducts a two-stage meta-analysis of O:E ratios using different input configurations.

```R
valmeta(measure="OE", O=n.events, E=e.events, N=n, slab=Study, data=EuroSCORE)
```

```R
valmeta(measure="OE", O=n.events, E=e.events, data=EuroSCORE)
```

```R
valmeta(measure="OE", Po=Po, Pe=Pe, N=n, data=EuroSCORE)
```

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

### Explore Fibrinogen Dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Fibrinogen.html

Basic commands to inspect the structure or visualize the loaded dataset.

```R
data(Fibrinogen)
## maybe str(Fibrinogen) ; plot(Fibrinogen) ...
```

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

### Estimate Generalizability

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/metapred.html

Use gen() to retrieve generalizability estimates. Setting genFUN to 0 returns all available estimates.

```R
gen(fit2, genFUN = 0)
```

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

### Load EuroSCORE data

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pm.html

Load the metamisc library and the EuroSCORE dataset for analysis.

```R
library(metamisc)
data(EuroSCORE)
```

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

### Load EuroSCORE dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/EuroSCORE.html

Load the EuroSCORE dataset into the R environment for analysis.

```R
data("EuroSCORE")
```

```R
data(EuroSCORE)
```

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

### data("impact")

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/impact.html

Loads the IMPACT dataset into the R environment.

```APIDOC
## Load Dataset

### Description
Loads the IMPACT dataset, which consists of 11,022 observations from 15 studies regarding traumatic brain injury.

### Usage
```r
data("impact")
```

### Variables
- **name** (string) - Name of the study
- **type** (string) - Type of study (RCT or OBS)
- **age** (numeric) - Age of the patient
- **motor_score** (numeric) - Glasgow Coma Scale motor score
- **pupil** (numeric) - Pupillary reactivity
- **ct** (numeric) - Marshall Computerized Tomography classification
- **hypox** (numeric) - Hypoxia (0=no, 1=yes)
- **hypots** (numeric) - Hypotension (0=no, 1=yes)
- **tsah** (numeric) - Traumatic subarachnoid hemorrhage (0=no, 1=yes)
- **edh** (numeric) - Epidural hematoma (0=no, 1=yes)
- **mort** (numeric) - 6-month mortality (0=alive, 1=dead)
```

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

### uvmeta()

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/index.html

Performs a univariate meta-analysis.

```APIDOC
## uvmeta()

### Description
Performs a univariate meta-analysis. Returns an object of class 'uvmeta'.
```

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

### Usage of stackedglm

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/stackedglm.html

Basic syntax for calling the stackedglm function to create a meta-model.

```R
stackedglm(models, family = binomial, data)
```

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

### Summarize uvmeta model

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/summary.uvmeta.html

Use this S3 method to generate summary statistics for a fitted univariate meta-analysis model object.

```R
# S3 method for class 'uvmeta'
summary(object, ...)
```

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

### Load DVTmodels dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/DVTmodels.html

Loads the DVTmodels dataset into the R environment for analysis.

```R
data(DVTmodels)
```

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

### Load Tzoulaki dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Tzoulaki.html

Load the dataset into the R environment for analysis.

```R
data("Tzoulaki")
```

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

### Load Collins dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Collins.html

Loads the Collins dataset into the R environment for analysis.

```R
data(Collins)
```

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

### Perform Bayesian one-stage random effects meta-analysis of the total O:E ratio

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/valmeta.html

Configures a Bayesian model with specific priors and handles missing N values by switching between Poisson and Binomial distributions.

```R
EuroSCORE.new <- EuroSCORE
EuroSCORE.new$n[c(1, 2, 5, 10, 20)] <-  NA
pars <- list(hp.tau.dist="dhalft",   # Prior for the between-study standard deviation
             hp.tau.sigma=1.5,       # Standard deviation for 'hp.tau.dist'
             hp.tau.df=3,            # Degrees of freedom for 'hp.tau.dist'
             hp.tau.max=10,          # Maximum value for the between-study standard deviation
             seed=5)                 # Set random seed for the simulations
fit3 <- valmeta(measure="OE", O=n.events, E=e.events, N=n, data=EuroSCORE.new,
        method="BAYES", slab=Study, pars=pars)
plot(fit3)
print(fit3$fit$model) # Inspect the JAGS model
print(fit3$fit$data)  # Inspect the JAGS data
```

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

### Perform frequentist random-effects meta-analysis with uvmeta

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/uvmeta.html

Executes a random-effects meta-analysis using the Roberts dataset and displays the summary and forest plot.

```R
data(Roberts)

# Frequentist random-effects meta-analysis
fit1 <- with(Roberts, uvmeta(r=SDM, r.se=SE, labels=rownames(Roberts)))
summary(fit1)
#> Call:
#> uvmeta.default(r = SDM, r.se = SE, labels = rownames(Roberts))
#> 
#> Random effects summary:	 0.36195 (SE: 0.0859)
#> Tau squared: 		 0.01322 (SE: 0.03431)
plot(fit1) #show a forest plot

fit1
#> Summary estimate with 95% confidence and (approximate) 95% prediction interval:
#> 
#>   Estimate        CIl        CIu        PIl        PIu 
#> 0.36194806 0.17636909 0.54752703 0.04923206 0.67466405 
```

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

### summary.riley(object, ...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/summary.riley.html

Computes summary estimates and confidence intervals for a riley object, assuming asymptotic normality for confidence intervals.

```APIDOC
## summary.riley(object, ...)

### Description
Provides the summary estimates of the alternative model for bivariate random-effects meta-analysis by Riley et al. (2008) with their corresponding standard errors. For confidence intervals, asymptotic normality is assumed.

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

### Arguments
- **object** (riley) - Required - A riley object.
- **...** (dots) - Optional - Arguments to be passed on to other functions (currently ignored).

### Value
Returns an array with confidence intervals for the estimated model parameters. For diagnostic test accuracy data, the resulting summary sensitivity and false positive rate are included.
```

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

### Explore Tzoulaki dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Tzoulaki.html

Basic commands to inspect the structure or visualize the dataset.

```R
data(Tzoulaki)
## maybe str(Tzoulaki) ; plot(Tzoulaki) ...
```

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

### Load Fibrinogen Dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Fibrinogen.html

Load the dataset into the R environment for analysis.

```R
data("Fibrinogen")
```

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

### Generate Forest Plot using Metafor

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pf.html

Creates a forest plot for overall survival using the metafor package.

```r
metafor::forest(x = Zhang$HR, ci.lb = Zhang$HR.025, ci.ub = Zhang$HR.975, slab = 
                  Zhang$Study, subset = (Zhang$outcome=="OS"),  xlab = 
                  "Hazard ratio of HER2 versus OS", refline = 1)
```

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

### Display Bayesian model object

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pm.html

Command to inspect the fitted Bayesian model object.

```R
fit
```

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

### Load Framingham dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Framingham.html

Load the Framingham dataset into the R environment.

```R
data("Framingham")
```

```R
data(Framingham)
```

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

### oecalc(OE, OE.se, OE.cilb, OE.ciub, OE.cilv, EO, EO.se, citl, citl.se, N, O, E, Po, Po.se, Pe, data, slab, add, g, level, ...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/oecalc.html

Calculates the ratio of total observed versus expected events with the corresponding sampling variance.

```APIDOC
## oecalc

### Description
This function calculates (transformed versions of) the ratio of total number of observed versus expected events with the corresponding sampling variance.

### Arguments
- **OE** (vector) - Optional - Estimated ratio of total observed versus total expected events.
- **OE.se** (vector) - Optional - Standard errors of the estimated O:E ratios.
- **OE.cilb** (vector) - Optional - Lower limits of the confidence interval for OE.
- **OE.ciub** (vector) - Optional - Upper limits of the confidence interval for OE.
- **OE.cilv** (vector) - Optional - Levels of confidence interval limits (default: 0.95).
- **EO** (vector) - Optional - Estimated ratio of total expected versus total observed events.
- **EO.se** (vector) - Optional - Standard errors of the estimated E:O ratios.
- **citl** (vector) - Optional - Estimated calibration-in-the-large statistics.
- **citl.se** (vector) - Optional - Standard error of the calibration-in-the-large statistics.
- **N** (vector) - Optional - Sample/group sizes.
- **O** (vector) - Optional - Total number of observed events.
- **E** (vector) - Optional - Total number of expected events.
- **Po** (vector) - Optional - (Cumulative) observed event probabilities.
- **Po.se** (vector) - Optional - Standard errors of Po.
- **Pe** (vector) - Optional - (Cumulative) expected event probabilities.
- **data** (data.frame) - Optional - Data frame containing the variables.
- **slab** (vector) - Optional - Labels for the studies.
- **add** (numeric) - Optional - Amount to add to zero counts (default: 1/2).
- **g** (string) - Optional - Function to transform estimates of the total O:E ratio.
- **level** (numeric) - Optional - Level for confidence interval (default: 0.95).

### Value
Returns an object of class c("mm_perf","data.frame") with columns:
- **theta**: The (transformed) O:E ratio.
- **theta.se**: Standard errors of the (transformed) O:E ratio.
- **theta.cilb**: Lower confidence interval of the (transformed) O:E ratios.
- **theta.ciub**: Upper confidence interval of the (transformed) O:E ratios.
- **theta.source**: Method used for calculating the (transformed) O:E ratio.
- **theta.se.source**: Method used for calculating the standard error of the (transformed) O:E ratio.
```

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

### riley(X, control)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/riley.html

Performs a meta-analysis on the provided dataset X. The control parameter allows for configuration of optimization settings such as maximum iterations.

```APIDOC
## riley(X, control)

### Description
Performs meta-analysis on a given dataset. This function is used to estimate parameters for surrogate markers, prognostic test studies, or other statistical data.

### Parameters
- **X** (data.frame) - Required - The dataset containing the study information to be analyzed.
- **control** (list) - Optional - A list of control parameters for the optimization process, such as 'maxit' for maximum iterations.

### Returns
- **fit** (object) - An object containing the estimated coefficients (beta1, beta2, psi1, psi2, rho) and model summary information.
```

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

### Calculate O:E ratio using oecalc

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/oecalc.html

Computes the O:E ratio and associated statistics from observed events, expected events, and total sample size.

```R
est1 <- oecalc(O = n.events, E = e.events, N = n, data = EuroSCORE, slab = Study)
est1
```

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

### Load Scheidler dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Scheidler.html

Load the Scheidler dataset into the R environment.

```R
data("Scheidler")
```

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

### Generate a Forest Plot for valmeta

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.valmeta.html

Basic usage of the plot function on a valmeta object to visualize performance estimates.

```R
data(EuroSCORE)
fit <- valmeta(cstat=c.index, cstat.se=se.c.index, cstat.cilb=c.index.95CIl,
               cstat.ciub=c.index.95CIu, N=n, O=n.events, data=EuroSCORE)
plot(fit)
```

```R
library(ggplot2)
plot(fit, theme=theme_grey())
```

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

### metapred(...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/metapred.html

Builds a generalizable model from data.

```APIDOC
## metapred(data, strata, formula, scope, family, ...)

### Description
Constructs a generalizable model from the provided dataset.

### Parameters
- **data** (data.frame) - Required - The dataset to use.
- **strata** (string) - Required - The variable name used for stratification.
- **formula** (formula) - Required - The model formula.
- **scope** (formula) - Optional - The scope for model selection.
- **family** (family) - Required - The error distribution and link function (e.g., binomial).
```

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

### Generate a forest plot with forest.default

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/forest.default.html

Use this S3 method to visualize meta-analysis results including study-specific effect sizes and summary estimates.

```R
# Default S3 method
forest(
  theta,
  theta.ci.lb,
  theta.ci.ub,
  theta.slab,
  theta.summary,
  theta.summary.ci.lb,
  theta.summary.ci.ub,
  theta.summary.pi.lb,
  theta.summary.pi.ub,
  title,
  sort = "asc",
  theme = theme_bw(),
  predint.linetype = 1,
  xlim,
  xlab = "",
  refline = 0,
  label.summary = "Summary Estimate",
  label.predint = "Prediction Interval",
  nrows.before.summary = 1,
  study.digits = 2,
  study.shape = 15,
  col.diamond = "white",
  col.predint = "black",
  size.study = 0.5,
  size.predint = 1,
  lty.ref = "dotted",
  ...
)
```

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

### Transform O:E ratio to log scale

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pm.html

Applies a natural logarithm transformation to the O:E ratio for meta-analysis preparation.

```R
EuroSCORE <- EuroSCORE %>%
    mutate(logoe = log(oe))
```

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

### uvmeta(r, r.se, r.vi, method, test, labels, na.action, n.chains, pars, verbose, ...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/uvmeta.html

Summarizes multiple estimates for a single parameter by assuming a fixed or random effects model across studies.

```APIDOC
## uvmeta(r, r.se, r.vi, method = "REML", test = "knha", labels, na.action, n.chains = 4, pars, verbose = FALSE, ...)

### Description
This function summarizes multiple estimates for a single parameter by assuming a fixed or random effects model across studies. It calculates a weighted mean accounting for sample size and between-study heterogeneity.

### Parameters
- **r** (numeric vector) - Required - Effect size of each study.
- **r.se** (numeric vector) - Optional - Standard error of the effect sizes.
- **r.vi** (numeric vector) - Optional - Sampling variance of the effect sizes.
- **method** (string) - Optional - Model type: "FE" for fixed-effect, or random-effects methods: "REML" (default), "DL", "HE", "SJ", "ML", "EB", "HS", "GENQ", "BAYES".
- **test** (string) - Optional - Method for test statistics and confidence intervals (default "knha").
- **labels** (character vector) - Optional - Labels for the studies.
- **na.action** (function) - Optional - Handling of NAs (default "na.fail").
- **n.chains** (numeric) - Optional - Number of chains for Gibbs sampler when method is "BAYES" (default 4).
- **pars** (list) - Optional - Additional configuration parameters for MCMC sampling and interval levels.
- **verbose** (boolean) - Optional - Display messages during fitting (default FALSE).
- **...** - Optional - Additional arguments passed to rma or runjags.
```

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

### Usage of dplot for mcmc.list

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/dplot.mcmc.list.html

Displays the S3 method signature for plotting posterior distributions from mcmc.list objects.

```R
# S3 method for class 'mcmc.list'
dplot(x, P, plot_type = "dens", ...)
```

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

### Obtain generalizability estimates

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/gen.html

Use these functions to retrieve generalizability estimates from a metapred model object. The ellipsis parameter allows passing arguments to subset.metapred for specific model selection.

```R
gen(object, ...)
generalizability(object, ...)
```

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

### Load Daniels dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/Daniels.html

Load the Daniels dataset into the R environment.

```R
data("Daniels")
```

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

### Obtain performance estimates using perf and performance

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/perf.html

Use these functions to retrieve performance metrics from a metapred object. By default, the final model is selected, but additional arguments can be passed to subset.metapred to target specific models.

```R
perf(object, ...)
performance(object, ...)
```

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

### Define oecalc function parameters

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/oecalc.html

Function signature for oecalc showing available arguments for calculating O:E ratios.

```R
oecalc(
  OE,
  OE.se,
  OE.cilb,
  OE.ciub,
  OE.cilv,
  EO,
  EO.se,
  citl,
  citl.se,
  N,
  O,
  E,
  Po,
  Po.se,
  Pe,
  data,
  slab,
  add = 1/2,
  g = NULL,
  level = 0.95,
  ...
)
```

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

### Calculate log of O:E ratio in R

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/oecalc.html

Computes the log of the O:E ratio using observed events, expected events, and total sample size.

```R
est2 <- oecalc(O = n.events, E = e.events, N = n, data = EuroSCORE, slab = Study, g = "log(OE)")
est2
```

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

### Perform one-stage meta-analysis of O:E ratios

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/valmeta.html

Conducts a one-stage meta-analysis using maximum likelihood estimation.

```R
valmeta(measure="OE", O=n.events, E=e.events, data=EuroSCORE, method="ML", 
        pars=list(model.oe="poisson/log"))
```

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

### inv.logit(x)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/inv.logit.html

Transforms a linear predictor into a probability.

```APIDOC
## inv.logit(x)

### Description
Transforms a linear predictor into a probability.

### Usage
inv.logit(x)

### Arguments
- **x** (numeric vector) - Required - A vector of numerics (between -Inf and Inf).

### Value
A vector of numerics between 0 and 1.
```

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

### Load IMPACT dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/impact.html

Loads the IMPACT dataset into the R environment.

```R
data("impact")
```

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

### Inspect the Zhang dataset

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pf.html

Displays the contents of the Zhang object, which includes study details, hazard ratios, and survival outcomes.

```R
Zhang
```

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

### plot(x, title, sort = "asc", xlim, refline, ...)

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/plot.riley.html

Generates a forest plot for each outcome of the bivariate meta-analysis.

```APIDOC
## plot(x, title, sort = "asc", xlim, refline, ...)

### Description
Generates a forest plot for each outcome of the bivariate meta-analysis based on the Riley et al. (2008) model.

### Arguments
- **x** (riley) - Required - An object of class 'riley'.
- **title** (string) - Optional - Title of the forest plot.
- **sort** (string) - Optional - Determines study ordering. Use "asc" for ascending effect size, "desc" for descending, or any other value to ignore ordering. Defaults to "asc".
- **xlim** (numeric) - Optional - The x limits (x1, x2) of the forest plot.
- **refline** (numeric) - Optional - A numeric value specifying a reference line.
- **...** (dots) - Optional - Additional parameters for generating forest plots.

### Example
```r
data(Scheidler)
fit <- riley(Scheidler[which(Scheidler$modality==1),])
plot(fit)
```
```

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

### Usage of dplot for valmeta objects

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/dplot.valmeta.html

Standard signature for the dplot S3 method.

```R
# S3 method for class 'valmeta'
dplot(x, par, distr_type, plot_type = "dens", ...)
```

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

### Load and Summarize IMPACT Data

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/impact.html

Load the dataset and generate summary statistics grouped by study name.

```R
data(impact)

by(impact, impact$name, summary)
```

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

### Usage of gelmanplot

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/gelmanplot.html

Basic syntax for calling the gelmanplot function.

```R
gelmanplot(...)
```

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

### Usage of rmplot for valmeta objects

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/rmplot.valmeta.html

Standard S3 method signature for plotting running means from a valmeta object.

```R
# S3 method for class 'valmeta'
rmplot(x, ...)
```

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

### Perform frequentist meta-analysis of c-statistics

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/articles/ma-pm.html

Uses the valmeta function with REML estimation and Hartung-Knapp-Sidik-Jonkman adjustments to summarize c-statistics.

```R
valmeta(cstat = c.index, cstat.se = se.c.index, cstat.cilb = c.index.95CIl, cstat.ciub = c.index.95CIu,
    N = n, O = n.events, method = "REML", test = "knha", slab = Study, data = EuroSCORE)
```

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

### Usage of acplot for Bayesian meta-analysis

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/acplot.valmeta.html

Displays the syntax for the S3 method acplot applied to a valmeta object.

```R
# S3 method for class 'valmeta'
acplot(x, ...)
```

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

### Usage of rmplot for uvmeta objects

Source: https://smartdata-analysis-and-statistics.github.io/metamisc/reference/rmplot.uvmeta.html

Standard S3 method signature for plotting running means from a uvmeta object.

```R
# S3 method for class 'uvmeta'
rmplot(x, ...)
```