gemtc (version 0.7-1)

mtc.network: Create an mtc.network

Description

Creates an object of class mtc.network

Usage

mtc.network(data.ab, treatments, description, data.re, data)
## S3 method for class 'mtc.network':
plot(x, layout=igraph::layout.circle, dynamic.edge.width=TRUE, ...)

Arguments

data.ab

Arm-level data. A data frame defining the arms of each study, containing the columns `study' and `treatment', where `treatment' must refer to an existing treatment ID if treatments were specified. Further columns define the data per arm, and depend on the

data.re

Relative effect data. A data frame defining the arms of each study, containing the columns `study' and `treatment', where `treatment' must refer to an existing treatment ID if treatments were specified. The column `diff' specifies the mean difference betw

treatments

Optional. A data frame with columns `id' and `description' defining the treatments or a vector giving the treatment IDs.

description

Optional. Short description of the network.

data

Deprecated. Arm-level data; automatically assigned to data.ab if it is not specified. Present for compatibility with older versions.

x

An mtc.network object.

layout

An igraph-compatible layout.

dynamic.edge.width

If set to TRUE, dynamically set the edge width based on the number of studies.

...

Additional arguments passed to plot.igraph.

Value

• For mtc.network, an object of the class mtc.network which is a list containing:
• description
  A short description of the network
• treatments
  A data frame describing the treatments
• data.ab
  A data frame containing the network data (arm-level)
• data.re
  A data frame containing the network data (relative effects)
• These are cleaned up and standardized versions of the arguments provided, or generated defaults for `treatments' if the argument was omitted.

encoding

utf8

Details

One-arm trials are automatically removed, which results in a warning.

Also see mtc.data.studyrow for a convenient way to import data from the one-study-per-row format, which is very popular for BUGS code.

See Also

mtc.data.studyrow mtc.model

Examples

# Create a new network by specifying all information.
treatments <- read.table(textConnection('
  id  description
  A   "Treatment A"
  B   "Treatment B"
  C   "Treatment C"'), header=TRUE)
data <- read.table(textConnection('
  study  treatment  responders  sampleSize
  01     A          2           100
  01     B          5           100
  02     B          6           110
  02     C          1           110
  03     A          3           60
  03     C          4           80
  03     B          7           80'), header=TRUE)
network <- mtc.network(data, description="Example", treatments=treatments)
plot(network)

# Create a new network by specifying only the data.
data <- read.table(textConnection('
  study  treatment  mean   std.dev  sampleSize
  01     A          -1.12  0.6      15
  01     B          -1.55  0.5      16
  02     A          -0.8   0.7      33
  02     B          -1.1   0.5      31'), header=TRUE)
network <- mtc.network(data)

# Print the network
print(network)
## MTC dataset: Network
##   study treatment  mean std.dev sampleSize
## 1     1         A -1.12     0.6         15
## 2     1         B -1.55     0.5         16
## 3     2         A -0.80     0.7         33
## 4     2         B -1.10     0.5         31