data-raw/case_prediction_data/case-predictors.R

In rrrlw/ripserr: Calculate Persistent Homology with Ripser-Based Engines

case_predictors <- data.frame()
case_predictors[27, 1] <- NA
stateList <- structure(list(name = c("Acre", "Alagoas", "Amap?", "Amazonas", 
                                     "Bahia", "Cear?", "Distrito Federal", "Esp?rito Santo", "Goi?s", "Maranhao", "Mato Grosso", 
                                     "Mato Grosso do Sul", "Minas Gerais", "Par?", "Para?ba", "Paran?", 
                                     "Pernambuco", "Piau?", "Rio de Janeiro", "Rio Grande do Norte", 
                                     "Rio Grande do Sul", "Rond?nia", "Roraima", "Santa Catarina", 
                                     "Sao Paulo", "Sergipe", "Tocantins"), abb = c("AC", "AL", "AP", 
                                                                                   "AM", "BA", "CE", "DF", "ES", "GO", "MA", "MT", "MS", "MG", "PA", "PB", 
                                                                                   "PR", "PE", "PI", "RJ", "RN", "RS", "RO", "RR", "SC", "SP", "SE", 
                                                                                   "TO")), row.names = c(15L, 17L, 26L, 21L, 10L, 23L, 8L, 9L, 18L, 
                                                                                                         11L, 5L, 7L, 19L, 22L, 3L, 20L, 16L, 6L, 24L, 1L, 12L, 25L, 2L, 
                                                                                         4L, 14L, 13L, 27L), class = "data.frame")


stateList <- stateList[order(stateList$name),]#sorted by state names
stateAbbSort <-sort(stateList$abb) #sorted vector of abbreviation

rownames(case_predictors) <- stateAbbSort

#----------------------Density = POP/Area----------------------#
Area <- read.csv("State Area 2020.csv", header = FALSE)
colnames(Area) <- c("State", "Area")

Population <- read.csv("Brazil Population 2013.csv", header = FALSE, fileEncoding="UTF-8-BOM")
colnames(Population) <- c("State", "Population")

curArea <- 0
curRow_A <- 1
curPop <- 0
curRow_P <- 1
pop_Density <- data.frame()

for(val in stateAbbSort){
  for(i in curRow_A : nrow(Area)){
    if(Area[i, 1] == val){
      curArea <- curArea + Area[i, 2]
    }else{
      curRow_A <- i
      break
    }
  }
  
  for(j in curRow_P : nrow(Population)){
    if(Population[j, 1] == val){
      curPop <- curPop + strtoi(Population[j,2], base = 0L)
    }else{
      curRow_P <- j
      break
    }
  }
  
  curDensity <- curPop/curArea
  
  pop_Density <- rbind(pop_Density, c(curArea, curPop, curDensity))
  curArea <- 0
  curPop <- 0
}

colnames(pop_Density) <- c("Area", "Population", "Density")
case_predictors[,1] <- pop_Density[,3]
colnames(case_predictors) <- "POP"

remove(pop_Density, Area, Population, curArea, curPop, curRow_A, curRow_P, curDensity)

#-----------------------------Temperature------------------------------#
Temperature <- read.csv("Temperature.csv")
removeRows <- vector()

for(i in 1 : nrow(Temperature)){ ##remove empty rows
  if(Temperature[i,6] == "#DIV/0!"){
    removeRows <- c(removeRows, c(i))
  }
}
Temperature <- Temperature[-removeRows,]

curTemp_Sum <- 0
curRow_T <- 1
TempAverage <- data.frame()

for(val in stateAbbSort){
  for(i in curRow_T : nrow(Temperature)){
    if(Temperature[i, 1] == val){
      curTemp_Sum <- curTemp_Sum + as.double(Temperature[i, 6])
      if(i == nrow(Temperature)){
        curTemp_Ave <- curTemp_Sum/(i - curRow_T)
        TempAverage <- rbind(TempAverage, c(curTemp_Ave))
      }
    }else{
      curTemp_Ave <- curTemp_Sum/(i - curRow_T)
      TempAverage <- rbind(TempAverage, c(curTemp_Ave))
      curRow_T <- i
      curTemp_Sum <- 0
      break
    }
  }
}
colnames(TempAverage) <- c("Average Temp")
rownames(TempAverage) <- stateAbbSort
case_predictors <- cbind(case_predictors, c(TempAverage[,1]))
colnames(case_predictors)[2] <- "TEMP"

remove(curTemp_Sum, curTemp_Ave, curRow_T, TempAverage, Temperature)

#------------------------Precipitation------------------------#
Precipitation <- read.csv("Precipitation.csv", fileEncoding = "latin1")
removeRows <- vector()

for(i in 1 : nrow(Precipitation)){ ##remove empty rows
  if(Precipitation[i,7] == "#DIV/0!"){
    removeRows <- c(removeRows, c(i))
  }
}
Precipitation <- Precipitation[-removeRows,]

curPrecip_Sum <- 0
curRow_P <- 1
PrecipAverage <- data.frame()

for(val in stateAbbSort){
  for(i in curRow_P : nrow(Precipitation)){
    if(Precipitation[i, 1] == val){
      curPrecip_Sum <- curPrecip_Sum + as.double(Precipitation[i, 7])
      if(i == nrow(Precipitation)){
        curPrecip_Ave <- curPrecip_Sum/(i - curRow_P)
        PrecipAverage <- rbind(PrecipAverage, c(curPrecip_Ave))
      }
    }else{
      curPrecip_Ave <- curPrecip_Sum/(i - curRow_P)
      PrecipAverage <- rbind(PrecipAverage, c(curPrecip_Ave))
      curRow_P <- i
      curPrecip_Sum <- 0
      break
    }
  }
}
colnames(PrecipAverage) <- c("Average Precip")
rownames(PrecipAverage) <- stateAbbSort
case_predictors <- cbind(case_predictors, c(PrecipAverage[,1]))
colnames(case_predictors)[3] <- "PRECIP"

remove(curPrecip_Sum, curPrecip_Ave, curRow_P, PrecipAverage, Precipitation)
#----------------------------Cases----------------------------#
Cases <- read.csv(("Dengue Cases 2013.csv"), header = FALSE)
case_predictors <- cbind(case_predictors, Cases[,2])
colnames(case_predictors)[4]<- "CASE"
remove(Cases, stateList)

usethis::use_data(case_predictors, overwrite = TRUE)