Demography package issue with aggregating data

Question

    # Function to construct a mortality demogdata object from HMD
hmd.mx <- function(country, username, password, label=country){
 path <- paste("https://www.mortality.org/hmd/", country, "/STATS/", "Mx_1x1.txt", sep = "")
  }
  userpwd <- paste(username, ":", password, sep = "")
  txt <- RCurl::getURL(path, userpwd = userpwd)
  con <- textConnection(txt)
  mx <- try(utils::read.table(con, skip = 2, header = TRUE, na.strings = "."),TRUE)
  close(con)
  if(class(mx)=="try-error")
    stop("Connection error at www.mortality.org. Please check username, password and country label.")

  path <- paste("https://www.mortality.org/hmd/", country, "/STATS/", "Exposures_1x1.txt", sep = "")
 userpwd <- paste(username, ":", password, sep = "")
  txt <- RCurl::getURL(path, userpwd = userpwd)
  con <- textConnection(txt)
  pop <- try(utils::read.table(con, skip = 2, header = TRUE, na.strings = "."),TRUE)
  close(con)
  if(class(pop)=="try-error")
    stop("Exposures file not found at www.mortality.org")
  obj <- list(type="mortality",label=label,lambda=0)
  obj$year <- sort(unique(mx[, 1]))
  #obj$year <- ts(obj$year, start=min(obj$year))
  n <- length(obj$year)
  m <- length(unique(mx[, 2]))
  obj$age <- mx[1:m, 2]
  obj$rate <- obj$pop <- list()
  for (i in 1:n.mort)
      { obj$rate[[i]] <- matrix(mx[, i + 2], nrow = m, ncol = n)
        obj$rate[[i]][obj$rate[[i]] < 0] <- NA
        obj$pop[[i]] <- matrix(pop[, i + 2], nrow = m, ncol = n)
        obj$pop[[i]][obj$pop[[i]] < 0] <- NA
        dimnames(obj$rate[[i]]) <- dimnames(obj$pop[[i]]) <- list(obj$age, obj$year)
       }
  names(obj$pop) = names(obj$rate) <- tolower(mnames)
  obj$age <- as.numeric(as.character(obj$age))
  if (is.na(obj$age[m])) {        
    obj$age[m] <- 2 * obj$age[m - 1] - obj$age[m - 2]            }
  return(structure(obj, class = "demogdata"))
}

Above is the code that we are using to import our population data into r .

NLdata <- hmd.mx(country = "NLD",username = "username",password="password")

This would be the specific code to obtain the Dutch data.

Would anyone happen to know how to add multiple countries into one, and put that data into one dataframe (same format as the demography data packages that we download)? So for example the mortality rates for the (Netherlands + France + Norway) / 3 into one package.

Answer 1

You can try this code. However I could not run your demography package. So you might need to edit the code a bit. Perhaps someone else can fill in the second part? I saw that no one has reacted yet.

C1 <- data.frame(Year = 1980:2018, value1 = rnorm(39), value2 = rnorm(39), Cat =rbinom(39,1,0.5), Country = "France")
C2 <- data.frame(Year = 1980:2018, value1 = rnorm(39), value2 = rnorm(39), Cat =rbinom(39,1,0.5),Country = "England")
C3 <- data.frame(Year = 1970:2018, value1 = rnorm(49), value2 = rnorm(49), Cat =rbinom(49,1,0.5),Country = "Netherlands")

C1 <- split(C1, C1$Cat)
C2 <- split(C2, C2$Cat)
C3 <- split(C3, C3$Cat)


list_all <- list(rbind(C1[[1]],C2[[1]],C3[[1]]),rbind(C1[[2]],C2[[2]],C3[[2]]))


Final_list <- lapply(list_all, function(x) x  %>% group_by(Year) %>% summarise(Val1 = mean(value1), Val2 = mean(value2), Country = "All") %>% as.data.frame)