Package index • csalert

Ensemble format and naming grammar

The ensemble of draws, the reporting-triangle input, and the grammar that names their columns.

csfmt_ensemble_v3(): Build a csfmt_ensemble_v3
validate_ensemble(): Check the shape of a csfmt_ensemble_v3
print(<csfmt_ensemble_v3>): Print a csfmt_ensemble_v3
csfmt_reporting_triangle_v3(): Build a csfmt_reporting_triangle_v3
reporting_triangle_matrix(): Turn a reporting triangle into one count matrix per series
isoyearweek_week_start(): The Monday that starts an ISO week
set_time_series_id(): Add a content-hash series id to a data.table
csfmt_var(): Build a measure column name from its parts
csfmt_parse(): Split a measure column name into its parts
csfmt_interpret(): Split every value column name of a table into its parts
q_label(): Write a probability as a quantile label
q_value(): Read the probability in a quantile label

Complete the weeks of a reporting triangle that are still being reported.

nowcast_delay_ecdf_v1(): Nowcast a reporting triangle from the delay pattern of settled weeks
nowcast_passthrough_to_ensemble_v1(): Build an ensemble from a reporting triangle with no nowcast

Stages that take an ensemble and return it with new columns, and the collapse to quantiles that ends the pipeline.

One ensemble that holds the draws in memory, or a table of percentiles to store, and the steps that take it in either state.

csfmt_ensemble_v4(): Build a csfmt_ensemble_v4
print(<csfmt_ensemble_v4>): Print a csfmt_ensemble_v4
csfmt_reporting_triangle_v4(): Build a csfmt_reporting_triangle_v4
nowcast_ecdf_v4(): Nowcast a reporting triangle from the delay pattern of settled weeks
nowcast_passthrough_v4(): Build an ensemble from a reporting triangle with no nowcast
ens_add_rate_v4(): Add a rate to a csfmt_ensemble_v4
short_term_trend_v4(): Fit the short-term trend of every draw of a csfmt_ensemble_v4
signal_detection_hlm_v4(): Flag the draws of a csfmt_ensemble_v4 above a historical limit
mem_thresholds_v4(): Classify the draws of a csfmt_ensemble_v4 into MEM intensity levels
ens_compress_v4(): Compress a csfmt_ensemble_v4 into a table to store
as_ensemble_v4(): Rebuild a csfmt_ensemble_v4 from a stored table

Replay a method on past dates and score it, and measure how fast counts arrive.

nowcast_censor(): Cut a reporting triangle back to what was known on a past date
nowcast_truth(): The settled total of each reference week
nowcast_backtest(): Replay a nowcast method on past as-of dates
nowcast_score_v1(): Score replayed nowcast quantiles against the settled truth
reporting_completion_v1(): Measure how fast the counts of a reporting triangle arrive
reporting_completion_trend_v1(): Reporting speed by year and by recent month, in one table

Check one input feed, and compare two runs of the pipeline.

qc_surveillance_data_v1(): Check that a surveillance feed has data and is up to date
qc_week_over_week_v1(): Compare this week's run with last week's run
compare_results(): Compare two collapsed result sets, column by column

The rolling-slope kernel, and the trend alarm for a surveillance sts object.

rolling_slope_matrix(): Rolling regression slope down every column of a matrix
short_term_trend_sts_v1(): Short-term trend alarms on a surveillance sts object

Simulate daily counts with known outbreaks and holiday effects.

simulate_baseline_data(): Simulate daily counts with no outbreak
simulate_seasonal_outbreak_data(): Add seasonal outbreaks to simulated daily counts
simulate_spike_outbreak_data(): Add short outbreaks to the end of simulated daily counts
add_holiday_effect(): Multiply simulated counts on public holidays