Short-term trend alarms on a surveillance sts object

Sets an alarm where a rolling quasi-Poisson regression finds a significant rise, in each column of a surveillance::sts object, after Benedetti (2019) doi:10.5588/pha.19.0002. It has not changed since 2024-06-24.

Usage

short_term_trend_sts_v1(sts, control = list(w = 5, alpha = 0.05))

Arguments

sts

A surveillance::sts object.

control

A list with w, the window width, and alpha, the significance level. A missing element takes its default.

Value

sts with its alarms set.

Details

Each time point gets the fit of observed ~ trend + log(population), by glm2::glm2(), on the w time points that end there. Its alarm is 1 when the slope is positive with a p-value below alpha, and 0 otherwise. The first w - 1 time points keep their alarm.

See also

vignette("csalert", package = "csalert"), which explains when to use each trend method. The csfmt_rts_data_v1 method of short_term_trend() is the table form of this function.

Other short-term trend functions: rolling_slope_matrix(), short_term_trend()

Examples

d <- cstidy::nor_covid19_icu_and_hospitalization_csfmt_rts_v1
d <- d[granularity_time=="isoyearweek"]
sts <- surveillance::sts(
  observed = d$hospitalization_with_covid19_as_primary_cause_n, # weekly counts
  start = c(d$isoyear[1], d$isoweek[1]), # the first week of the series
  frequency = 52
)
x <- csalert::short_term_trend_sts_v1(
  sts,
  control = list(
    w = 5,
    alpha = 0.05
  )
)
plot(x)