Count number and proportion of observations in equal intervals

Usage

adcp_count_obs(
  dat,
  column = sea_water_speed_cm_s,
  n_ints = 12,
  n_digits = 4,
  label_sep = "\n",
  lowest = FALSE
)

Arguments

dat: Data frame including the column with observations that will be assigned to intervals. The interval size is determined by the range of these observations and n_ints.
column: Column in dat that will be assigned to intervals for frequency table (NOT QUOTED).
n_ints: Number of intervals to divide observations into. The interval size is determined by the range of these observations and n_ints. Alternatively, a vector of two or more unique break points. Passed to the breaks argument of cut().
n_digits: Number of digits to show in interval labels. Note that additional digits may be used in the actual break values. See vignettes for more detail. Passed to the dig.lab argument of cut().
label_sep: Separator for the interval labels ("lower to upper"). Default is a new line to save room on plot axis.
lowest: Logical. If TRUE, the first interval will be left *and* right inclusive. Passed to the include.lowest argument of cut().

Value

Returns a data frame of lower and upper interval limits (right inclusive), and frequency and proportion of observations in each interval.

Details

Assigns intervals to column using the cut() function. Argument right = TRUE to match the intervals assigned in the openair::windRose() function (which is called in adcp_plot_current_rose()). This means intervals are right-inclusive, i.e., a value of 4 is assigned to the interval (1,4].

NOTE: The openair::windRose() function has the cut() argument include.lower hard-coded to FALSE. This causes an issue with the intervals extracted from adcp_count_obs() being passed to adcp_plot_current_rose() (which passes all arguments to openair::windRose()).

Assume the first interval is in the form (x1, x2], and the minimum value of the observations is xmin.

This means that if the round(x1, digits = n_digits) equals xmin, values of xmin will NOT be assigned to an interval. In this case, xmin falls outside of the first interval (because it is left exclusive). To avoid this, n_digits should be large enough to ensure round(x1, digits = n_digits) < xmin. This could also be solved if include.lower could be set to TRUE in openair::windRose().

NOTE: I submitted a Pull Request to Open Air. I think it was merged. So could update the include.lower call here, but will need to think about this.