Package: scanstatistics 1.1.1

Paul Romer Present

scanstatistics: Space-Time Anomaly Detection using Scan Statistics

Detection of anomalous space-time clusters using the scan statistics methodology. Focuses on prospective surveillance of data streams, scanning for clusters with ongoing anomalies. Hypothesis testing is made possible by Monte Carlo simulation. Allévius (2018) <doi:10.21105/joss.00515>.

Authors:Benjamin Allévius [aut], Paul Romer Present [ctb, cre]

scanstatistics_1.1.1.tar.gz
scanstatistics_1.1.1.zip(r-4.7-x86_64)scanstatistics_1.1.1.zip(r-4.7-arm64)scanstatistics_1.1.1.zip(r-4.6-x86_64)scanstatistics_1.1.1.zip(r-4.6-arm64)
scanstatistics_1.1.1.tgz(r-4.6-x86_64)scanstatistics_1.1.1.tgz(r-4.6-arm64)
scanstatistics_1.1.1.tar.gz(r-4.7-arm64)scanstatistics_1.1.1.tar.gz(r-4.7-x86_64)scanstatistics_1.1.1.tar.gz(r-4.6-arm64)scanstatistics_1.1.1.tar.gz(r-4.6-x86_64)
scanstatistics_1.1.1.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
scanstatistics/json (API)

# Install 'scanstatistics' in R:

install.packages('scanstatistics', repos = c('https://promerpr.r-universe.dev', 'https://cloud.r-project.org'))

Bug tracker:https://github.com/promerpr/scanstatistics/issues

Uses libs:

c++– GNU Standard C++ Library v3

Datasets:

NM_geo - Longitude and latitude of New Mexico county seats.
NM_map - Data to plot the counties of New Mexico.
NM_popcas - Population and brain cancer cases in New Mexico counties during 1973-1991.

On CRAN:

cpp

5.18 score 2 stars 51 scripts 522 downloads 23 exports 29 dependencies

Last updated from:d3f0113c62. Checks:10 NOTE, 2 OK, 3 FAIL. Indexed: yes.

Target	Result	Time
linux-devel-arm64	NOTE	198
linux-devel-x86_64	NOTE	213
source / vignettes	OK	259
linux-release-arm64	NOTE	186
linux-release-x86_64	NOTE	237
macos-release-arm64	NOTE	190
macos-release-x86_64	NOTE	486
macos-oldrel-arm64	FAIL	123
macos-oldrel-x86_64	FAIL	246
windows-devel-arm64	NOTE	425
windows-devel-x86_64	NOTE	243
windows-release-arm64	NOTE	392
windows-release-x86_64	NOTE	186
windows-oldrel-x86_64	FAIL	53
wasm-release	OK	149

Exports:coords_to_knn df_to_matrix dist_to_knn flexible_zones get_zero_indices get_zone gumbel_pvalue knn_zones mc_pvalue scan_bayes_negbin scan_bayes_negbin_cpp scan_eb_negbin scan_eb_negbin_cpp scan_eb_poisson scan_eb_poisson_cpp scan_eb_zip scan_eb_zip_cpp scan_pb_perm_cpp scan_pb_poisson scan_pb_poisson_cpp scan_permutation score_locations top_clusters

Dependencies:cli cpp11 dplyr generics glue ismev lattice lifecycle magrittr Matrix mgcv nlme pillar pkgconfig plyr purrr R6 Rcpp RcppArmadillo rlang sets stringi stringr tibble tidyr tidyselect utf8 vctrs withr

Introduction to scanstatistics

Rendered fromintroduction.Rmdusingknitr::rmarkdown

Last update: 2022-10-24
Started: 2016-09-27

Citation

Development and contributors

Readme and manuals

Help Manual

Help page	Topics
Get the k nearest neighbors for each location, given its coordinates.	coords_to_knn
Convert a long data frame to a wide matrix.	df_to_matrix
Given a distance matrix, find the k nearest neighbors.	dist_to_knn
Computes the flexibly shaped zones as in Tango (2005).	flexible_zones
Extract a zone from the set of all zones.	get_zone
Calculate the Gumbel p-value for a scan statistic.	gumbel_pvalue
Find the increasing subsets of k nearest neighbors for all locations.	knn_zones
Calculate the Monte Carlo p-value for a scan statistic.	mc_pvalue
Longitude and latitude of New Mexico county seats.	NM_geo
Data to plot the counties of New Mexico.	NM_map
Population and brain cancer cases in New Mexico counties during 1973-1991.	NM_popcas
Calculate the negative binomial bayesian scan statistic..	scan_bayes_negbin
Calculate the expectation-based negative binomial scan statistic.	scan_eb_negbin
Calculate the expectation-based Poisson scan statistic.	scan_eb_poisson
Calculate the expectation-based ZIP scan statistic.	scan_eb_zip
Calculate the population-based Poisson scan statistic.	scan_pb_poisson
Calculate the space-time permutation scan statistic.	scan_permutation
scanstatistics: Space-time anomaly detection using scan statistics.	scanstatistics
Score each location over zones and duration.	score_locations
Get the top (non-overlappig) clusters.	top_clusters