-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathreadme.Rmd
69 lines (45 loc) · 3.54 KB
/
readme.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
---
title: "bmat"
output: github_document
---
## Table of contents
[Summary](#summary) \
[Required software](#reqs) \
[Installation](#install) \
[Data inputs](#data) \
[Use cases](#use) \
## <a name="summary"></a>
## Summary
The bmat package includes global and one-country implementations of the BMis and BMat model for the estimation of the maternal mortality rate (MMR) defined as the number of maternal deaths in a population per 100,000 live births
## <a name="reqs"></a>
## Requirements
For usage, an R installation (4.2.2), a JAGS installation (4.x), and a working developer environment such as an Rtools installation (4.2) are required. R and Rtools can be downloaded on CRAN. JAGS is a program for the analysis of Bayesian models using Markov Chain Monte Carlo (MCMC)
(Plummer, 2017). JAGS is written in C++ and is portable to all major operating systems. JAGS
is available for download [here](https://sourceforge.net/projects/mcmc-jags/).
Developer environment options.
Windows: Install Rtools.
Mac: Install Xcode from the Mac App Store.
Linux: Install a compiler and various development libraries (details vary across different flavors of Linux).
## <a name="install"></a>
## Installation
1. Obtain the version of bmat you wish to use. For the latest version clone the master branch of this repository. For versions of bmat which are associated with release dates visit the release dates and tags pages on github.
2. Create project from existing directory in R studio. Use the folder obtained from step 1 as the existing directory.
3. Enter devtools::install() in Rstudio console while having the project open in Rstudio.
4. When prompted to install required packages, enter "yes" in the console.
Package dependencies are listed in the package
DESCRIPTION file and will be automatically installed upon installing the main package.
Versions of bmat which are associated with release dates may also be installed from binary using the base R function install.package().
Note: Developers may wish to forgo installation and use the devtools::load_all() function to actively make changes to package code without iterative installation.
## <a name="data"></a>
## Data inputs
The data for the use cases can be found here [bmat/data-raw/](https://github.com/WorldHealthOrganization/bmat/tree/master/data-raw). A meta file describing the data for use cases can be found here. [bmat/data-raw/meta_data_bmat_gather.xlsx](https://github.com/WorldHealthOrganization/bmat/blob/master/data-raw/meta_data_bmat_gather.xlsx). For more detailed information about the data used please refer to the WHO report, Trends in maternal mortality: 2000 to 2020. Geneva: WHO, UNICEF, UNFPA and The World Bank; 2023.
## <a name="use"></a>
## Use cases
#### Case 1. One country run
To produce one country estimates which are based on the latest fixed global estimates, refer to the following vignette [bmat/vignettes/0_fit_one_country.R](https://github.com/WorldHealthOrganization/bmat/blob/master/vignettes/0_fit_one_country.R)
#### Case 2. Full estimation process
For the full process of producing estimates for all countries from start to finish, run the following vignettes in order. Vignettes can be found here [bmat/vignettes/](https://github.com/WorldHealthOrganization/bmat/tree/master/vignettes). The second and third use case are computationally intensive when using settings that are beyond that of a test run. This will generally require server computation.
1_process_data.R \
2_fit_models.R \
3_aggregates.R \
4_coutnry_profile_loop.R \