strs: Stratified Random Sampling

In sollano/forestmangr: Forest Mensuration and Management

Stratified Random Sampling

Description

Function for processing forest inventory data using stratified random sampling.

Usage

strs(
  df,
  Yi,
  plot_area,
  strata_area,
  strata,
  m3ha = FALSE,
  .groups = NA,
  age = NA,
  alpha = 0.05,
  error = 10,
  dec_places = 4,
  pop = "inf",
  tidy = TRUE
)

Arguments

df	a data frame.
Yi	Quoted name of the volume variable, or other variable one desires to evaluate, in quotes.
plot_area	Quoted name of the plot area variable, or a numeric vector with the plot area value. The plot area value must be in square meters.
strata_area	Quoted name of the strata area variable, or a numeric vector with the plot strata values. If there are more than 1 area values, it's possible to use a vector with all area values, like so:c(14.4, 16.4, 14.2). The strata area values must be in hectares.
strata	Quoted name of the subdivision variable(s), also known as strata. If this argument is not supplied, the defined groups in the data frame will be used, if they exist.
m3ha	Boolean value. If TRUE Yi variable is treated in m3/ha, else, in m3. Default: FALSE.
.groups	Optional argument. Quoted name(s) of additional grouping variable(s) that, if supplied, will be used to run multiple surveys, one for each level. If this argument is NA, the defined groups in the data frame will be used, if they exist. Default: NA.
age	Optional parameter. Quoted name of the age variable. Calculates the average age supplied. NA.
alpha	Numeric value for the significance value used in the t-student estimation. Default: 0.05.
error	Numeric value for the minimum admitted error value in the survey, in percentage. Default: 10.
dec_places	Numeric value for the number of decimal places to be used in the output tables. Default: 4.
pop	Character value for the type of population considered in the calculations. This can be either infinite ("inf") or finite ("fin"). Default: "inf".
tidy	Boolean value that defines if the output tables should be tidied up or not. Default: TRUE.

Details

This function allows the user to processes inventory data using stratified random sampling for n forest subdivisions (strata), for finite or infinite populations. It's possible to run multiple sampling analysis using a factor variable indicated in the .groups() parameter.

Value

A list containing two data frames, one with information for each strata, and one with the stratified sampling results.

Author(s)

Sollano Rabelo Braga [e-mail]

References

Campos, J. C. C. and Leite, H. G. (2017) Mensuracao Florestal: Perguntas e Respostas. 5a. Vicosa: UFV.

Soares, C. P. B., Paula Neto, F. and Souza, A. L. (2012) Dendrometria e Inventario Florestal. 2nd ed. Vicosa: UFV.

See Also

other sampling functions: sprs for Simple Random Sampling, and ss_diffs for Systematic Sampling.

Examples

library(forestmangr)
data("exfm1")
data("exfm2")
data("exfm6")

# The objective is to sample an area, with an error of 5%.
# First we run a pilot inventory, considering a 5% error and a finite population:
head(exfm1)

strs(exfm1, "VWB", "PLOT_AREA", "STRATA_AREA", strata = "STRATA", error = 5, pop = "fin")

# With these results, in order to meet the desired error of 5%, we'll need to sample 24 more plots,
# 4 in stratum 1, 8 in stratum 2, and 12 in stratum 3.
# After sampling the necessary plots, we now run a definitive inventory,
# considering an 5% error and a finite population:
exfm2

strs(exfm2, "VWB", "PLOT_AREA", "STRATA_AREA", strata = "STRATA", error = 5, pop = "fin")

# The desired error value was met.

# Area values can be numeric:
strs(exfm2, "VWB", 1000, c(14.4, 16.4,14.2), strata = "STRATA", error = 5, pop = "fin")

# Optional variable age, and one stratified sampled inventory for each map:
exfm6

strs(exfm6, "VWB", "PLOT_AREA", "STRATA_AREA", strata ="STRATA", .groups = "MAP", age = "AGE")

sollano/forestmangr documentation built on Dec. 3, 2024, 6:51 p.m.