Summary tables total for encounter rate CV differ for dht versus dht2

[LHMarshall]

The summary tables in the output have different total encounter rate CV's for dht versus dht2. Note the variability surrounding the estimates is the same for dht and dht2. It is possible this is related to the inconsistent ways things are handled in the summary tables.

Simulated example to compare dht results with dht2 for different ER variance estimators R2, O2 and S2

R2 comparison

O2 comparison

S2 comparison

It is now as consistent as the others with only the total ER cv looking to have an issue

Simulation Code

library(dsims)
outer1 <- matrix(c(0,0,1000,0,1000,500,0,500,0,0),ncol=2, byrow=TRUE)
outer2 <- matrix(c(0,500,1000,500,1000,1000,0,1000,0,500),ncol=2, byrow=TRUE)
outer3 <- matrix(c(0,1000,1000,1000,1000,1500,0,1500,0,1000),ncol=2, byrow=TRUE)
pol1 <- sf::st_polygon(list(outer1))
pol2 <- sf::st_polygon(list(outer2))
pol3 <- sf::st_polygon(list(outer3))
sfc <- sf::st_sfc(pol1,pol2,pol3)
strata.names <- c("South", "central", "North")
mp1 <- sf::st_sf(strata = strata.names, geom = sfc)

region <- make.region(region.name = "study.area", 
                      strata.name = strata.names, 
                      shape = mp1)
plot(region)

design <- make.design(region = region, 
                      samplers = rep(10,3),
                      truncation = 10)

pop.desc <- make.population.description(region = region,
                                        density = make.density(region= region),
                                        N = c(10, 250, 250))

detect <- make.detectability(scale.param = 5,
                             truncation = 10)

sim <- make.simulation(reps = 10,
                       design = design,
                       population.description = pop.desc,
                       detectability = detect)

set.seed(666)
survey <- run.survey(sim)
plot(survey)

eg.data <- survey@dist.data

fitR2 <- ds(eg.data,
            key = "hn",
            nadj = 0,
            er_var = "R2",
            truncation = 10)
fitR2.dht <- fitR2$dht

fitO2 <- ds(eg.data,
          key = "hn",
          nadj = 0,
          er_var = "O2",
          truncation = 10)
fitO2.dht <- fitO2$dht

fitS2 <- ds(eg.data,
            key = "hn",
            nadj = 0,
            er_var = "S2",
            truncation = 10)
fitS2.dht <- fitS2$dht

# Now use dht2 to check the results

dht2.R2 <- dht2(fitR2.ddf,
                flatfile = eg.data,
                strat_formula = ~Region.Label,
                er_est = "R2",
                stratification = "geographical")

dht2.O2 <- dht2(fitR2.ddf,
                flatfile = eg.data,
                strat_formula = ~Region.Label,
                er_est = "O2",
                stratification = "geographical")

dht2.S2 <- dht2(fitR2.ddf,
                flatfile = eg.data,
                strat_formula = ~Region.Label,
                er_est = "S2",
                stratification = "geographical")

Originally posted by @LHMarshall in #174