Skip to contents

performance_map_report

Source: R/performance_map_report.R
performance_map_report.Rd

Hierarchical clustering of performances, using audience crossover data to calculate the effective "distance" between performances.

  • Distance is calculated using Simpson's distance metric

  • Distance is corrected by time using a binomial regression on time between performances

  • Clustering is hierarchical using Ward D2 distance

  • Mapping is classic multidimensional scaling, i.e. PCA

Usage

performance_map_report

# S3 method for class 'performance_map'
read(
  report,
  since = Sys.Date() - 365 * 5,
  until = Sys.Date() + 365,
  filter_expr = NULL,
  ...
)

# S3 method for class 'performance_map'
process(report, n_clusters = 8, ...)

# S3 method for class 'performance_map'
write(report, n_clusters = 8, highlight_since = Sys.Date() - 365, ...)

Format

An object of class performance_map (inherits from email_report, report, list) of length 0.

Arguments

report

report object

since

POSIXct performance data on/after this date will be returned

until

POSIXct performance data on/before this date will be returned

filter_expr

expression to use when filtering performances, i.e. read_tessi("performances") %>% filter(!!filter_expr)

...

not used

n_clusters

integer number of clusters to make

highlight_since

POSIXct performances on/after this date will be highlighted in the returned plots

Functions

  • read(performance_map): read data for performance map

  • process(performance_map): analyze and cluster performances

  • write(performance_map): write pdf of performance maps and an file for importing into Tessitura