|
| 1 | +Nowcaster |
| 2 | +================ |
| 3 | + |
| 4 | +<a href='https://github.com/covid19br/nowcaster'><img src='man/figures/nowcaster.png' align="right" width="140" /></a> |
| 5 | +<a href='https://github.com/covid19br/nowcaster'><img src='man/figures/nowcaster_rev.png' align="right" width="140" /></a> |
| 6 | + |
| 7 | +<!-- badges: start --> |
| 8 | + |
| 9 | +<!-- [](https://cran.r-project.org/web/checks/check_results_nowcaster.html) --> |
| 10 | + |
| 11 | +<!-- [](https://cran.r-project.org/package=nowcaster) --> |
| 12 | + |
| 13 | +[](https://github.com/nowcaster) |
| 14 | +[-blue.svg)](https://github.com/covid19br/nowcaster/blob/main/LICENSE.md) |
| 16 | +[](https://lifecycle.r-lib.org/articles/stages.html#experimental) |
| 17 | +<!-- badges: end --> |
| 18 | + |
| 19 | +`nowcaster` is an R package for “nowcasting” epidemiological time-series |
| 20 | +on individual level data. |
| 21 | + |
| 22 | +Every single system of notification has an intrinsic delay between the |
| 23 | +`date of onset` of the event and the `date of report`. `nowcaster` can |
| 24 | +estimate how many counts of any epidemiological data of interest |
| 25 | +(*i.e.*, daily cases and deaths counts) by fitting a negative binomial |
| 26 | +model to the time steps of delay between onset date of the event, |
| 27 | +(*i.e.*, date of first symptoms for cases or date of occurrence of |
| 28 | +death) and the date of report (*i.e.*, date of notification of the case |
| 29 | +or death). |
| 30 | + |
| 31 | +## Installing |
| 32 | + |
| 33 | +After have a proper `INLA` installation to install `nowcaster` package |
| 34 | +simply run the code below in R: |
| 35 | + |
| 36 | +``` r |
| 37 | +devtools::install_github("https://github.com/covid19br/nowcaster") |
| 38 | +``` |
| 39 | + |
| 40 | +If you have any problem installing, please refer to next section on the |
| 41 | +dependencies of the package. |
| 42 | + |
| 43 | +## Dependencies |
| 44 | + |
| 45 | +`nowcaster` is based on the |
| 46 | +[`R-INLA`](https://www.r-inla.org/download-install) and |
| 47 | +[`INLA`](https://inla.r-inla-download.org/r-inla.org/doc/inla-manual/inla-manual.pdf) |
| 48 | +packages for “**I**ntegrated **N**ested **L**aplace **A**pproximation” |
| 49 | +algorithm to Bayesian inference. `INLA` is a fast alternative to others |
| 50 | +methods for Bayesian inference like **MCMC**. An introduction to `INLA` |
| 51 | +can be found |
| 52 | +[here](https://becarioprecario.bitbucket.io/inla-gitbook/index.html). |
| 53 | + |
| 54 | +`nowcaster` it was built for epidemiological emergency use, it was |
| 55 | +constructed for the Brazilian Severe Acute Respiratory Illness (SARI) |
| 56 | +surveillance system (SIVEP-Gripe), at the time of Covid-19 pandemic. |
| 57 | + |
| 58 | +Before installing the package certify you have an active installation of |
| 59 | +`INLA`, to do so you can run the following code: |
| 60 | + |
| 61 | +``` r |
| 62 | +install.packages("INLA", |
| 63 | + repos=c(getOption("repos"), |
| 64 | + INLA="https://inla.r-inla-download.org/R/stable"), |
| 65 | + dep=TRUE) |
| 66 | +``` |
| 67 | + |
| 68 | +If you want more detail on other possible installations of `INLA`, |
| 69 | +please refer to the official |
| 70 | +[page](https://www.r-inla.org/download-install) of the package. |
| 71 | + |
| 72 | +## Similar Initiatives |
| 73 | + |
| 74 | +There are other alternative packages, that can produce nowcasting |
| 75 | +estimation, here it is some options: |
| 76 | + |
| 77 | +- [Epinowcast](https://package.epinowcast.org/) |
| 78 | +- [Surveillance](https://surveillance.r-forge.r-project.org/pkgdown/index.html) |
0 commit comments