I am a bit stuck.
I have age startified risk ratios from published data from which i have sampled 10,000 iterations. I have a population census data of the population that i am going to apply the data to.
I effectively want to calculate an overall age standardised risk ratio weighted by the population proportion?
I have provided sample data below
structure(list(age = c("20-29", "30-39", "40-49"), pop_prop = c("0.2",
"0.3", "0.5"), smpls = list(c("0.549887166573646", "1.06662760919436",
"1.07106187160684", "0.853150943091018", "0.219922669374338"),
c("1.00205787604554", "0.73378506481059", "1.25277379884288",
"0.400571117409361", "0.832445422358434"), c("0.559245921265836",
"0.9315495664855", "0.453168951982812", "-0.830177080275652",
"0.459043679345258"))), row.names = c(NA, -3L), class = c("grouped_df",
"tbl_df", "tbl", "data.frame"), groups = structure(list(age = c("20-29",
"30-39", "40-49"), .rows = list(1L, 2L, 3L)), row.names = c(NA,
-3L), class = c("tbl_df", "tbl", "data.frame"), .drop = TRUE))
Each row contains the age group, the population proportion and the nested variable smples which represents the log (RR)
I want to derive interations for an overall age standardised RR weighted by the population proportion.
Can i effectively mutiply the pop_prop * smpls and then add them up?
