README.md

In sangeetabhatia03/promedr: Produce clean incidence time series from ProMED and HealthMap data

promedr

The goal of promedr is to pre-process data from ProMED and HealthMap and produce a clean incidence time series that can be used in analysis downstream.

ProMED/HealthMap data are made available as a CSV with the following columns:

• Location
• Country
• Region/City
• Disease
• Species
• Feed Language
• HM Alert ID
• Headline
• URL
• Issue Date
• Alert Tag
• New SC
• New SD
• New CC
• New CS
• Cumulative SC
• Cumulative SD
• Cumulative CC
• RO
• Feed Name
• Report Source - Primary
• Secondary Report Source

Processing this data consists of the following steps:

• Extract the total cumulative case count as a sum of desired case categories (Confirmed, Suspected, or both)
• If there are any duplicate alerts for a given day, merge them according to pre-specified set of rules.
• Any inconsistent data points at the end of the time series i.e., those that make the cumulative case count non-increasing, are moved using Chebyshev’s inequality.
• Any points in the middle of the time series that cause the cumulative case count to be decreasing, are removed using pre-specified rules.
• Finally, missing data are imputed.

Installation

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("sangeetabhatia03/promedr")

sangeetabhatia03/promedr documentation built on March 12, 2020, 7:25 a.m.