revisit_bibd: Create a balanced incomplete block panel revisit design
Description
Create a revisit design for panels in a survey that specifies the time
periods for the units of each panel to be sampled based on searching for a
D-optimal block design that is a member of the class of generalized Youden
designs. The resulting design need not be a balanced incomplete block
design. Based on algorithmic idea by Cook and Nachtsheim (1989) and
implemented by Robert Wheeler.
Usage
revisit_bibd(
n_period,
n_pnl,
n_visit,
nsamp,
panel_name = "BIB",
begin = 1,
skip = 1,
iter = 30
)
Value
A two-dimensional array of sample sizes to be sampled for each panel
and each sampling occasion.
Arguments
n_period
Number of time periods for the survey design. Typically,
number of periods if sampling occurs once per period or number of months if
sampling occurs once per month. (v, number of varieties/treatments in BIBD
terms)
n_pnl
Number of panels (b, number of blocks in BIBD terms)
n_visit
Number of time periods to be visited in a panel (k, block
size in BIBD terms)
nsamp
Number of samples in each panel.
panel_name
Prefix for name of each panel
begin
Numeric name of first sampling occasion, e.g. a specific
period.
skip
Number of sampling occasions to skip between planned sampling
periods, e.g., sampling will occur only every 5 periods if skip = 5.
iter
Maximum number of iterations in search for D-optimal
Generalized Youden Design.
The function uses find.BIB function from crossdes package to
search for a D-optimal block design. crossdes uses package AlgDesign
to search balanced incomplete block designs.
References
Cook R. D. and C. Nachtsheim. (1989). Computer-aided blocking of factorial
and response-surface designs. Technometrics31(3), 339-346.
See Also
revisit_dsgn
to create a panel revisit design
revisit_rand
to create a panel revisit design with random
assignment to panels and time periods
pd_summary
to summarize characteristics of a
panel revisit design
# Balanced incomplete block design with 20 sample occasions, 20 panels,# 3 visits to each unit, and 20 units in each panel.revisit_bibd(n_period = 20, n_pnl = 20, n_visit = 3, nsamp = 20)