39 lines
2.4 KiB
R
39 lines
2.4 KiB
R
#### NEXT STEPS!!!! USE CORRELATION TO DRAW FROM EACH MIGRANT IN A GIVEN YEAR
|
|
##########################Model Migration Trends
|
|
library(tidyverse)
|
|
library(fixest)
|
|
library(corrplot)
|
|
######Checking correlations with migration rates
|
|
DEMOGRAPHIC_DATA <- readRDS("Data/Cleaned_Data/Wyoming_County_Demographic_Data.Rds")
|
|
|
|
#Extract the population trend data to connect with demographics (Population,births,deaths)
|
|
POP_DATA <- readRDS("Data/Cleaned_Data/Wyoming_County_Population.Rds")
|
|
#Identify births, deaths an migration from existing data.
|
|
DEMO1 <- DEMOGRAPHIC_DATA
|
|
DEMO2 <- DEMOGRAPHIC_DATA %>% mutate(Year=Year+1,Age=Age+1) %>% rename(PREV_MALE=Num_Male,PREV_FEMALE=Num_Female)
|
|
DEMO_DATA <- inner_join(DEMO1,DEMO2) %>% mutate(Male=Num_Male-PREV_MALE,Female=Num_Female-PREV_FEMALE) %>% select(County,Year,Age,Male,Female) %>% arrange(County,Year,Age)
|
|
COR_MAT_DATA_FULL <- pivot_wider(DEMO_DATA,values_from=c(Male,Female),names_from=Age)
|
|
COR_MAT_DATA_FULL <- POP_DATA %>% left_join(COR_MAT_DATA_FULL )
|
|
COR_DATA <- COR_MAT_DATA_FULL %>% filter(Year>2010) %>% select(-County,-Year,-Births,-Deaths,-Population)
|
|
COR <- cor(COR_DATA,use="pairwise.complete.obs")
|
|
COR_RES <- COR["Migration",2:(ncol(COR))]
|
|
COR_RES <- cbind(rep(1:90,2),c(rep("Male",ncol(COR)/2),rep("Female",ncol(COR)/2)),as.numeric(COR_RES)) %>% as_tibble
|
|
colnames(COR_RES) <- c("Age","Sex","Cor")
|
|
COR_RES <- COR_RES %>% mutate(Age=as.integer(Age),Cor=as.numeric(Cor))
|
|
ggplot(COR_RES,aes(x=Age,y=Cor,group=Sex,color=Sex))+geom_smooth(span=0.25)+geom_point()
|
|
########################Combine Male and Female Since they look similar
|
|
DEMO_DATA <- inner_join(DEMO1,DEMO2) %>% mutate(Male=Num_Male-PREV_MALE,Female=Num_Female-PREV_FEMALE,Change=Male+Female) %>% select(County,Year,Age,Change) %>% arrange(County,Year,Age)
|
|
COR_MAT_DATA_FULL <- pivot_wider(DEMO_DATA,values_from=c(Change),names_from=Age)
|
|
COR_MAT_DATA_FULL <- POP_DATA %>% left_join(COR_MAT_DATA_FULL )
|
|
COR_DATA <- COR_MAT_DATA_FULL %>% filter(Year>2010) %>% select(-County,-Year,-Births,-Deaths,-Population)
|
|
COR <- cor(COR_DATA,use="pairwise.complete.obs")
|
|
COR_RES <- COR["Migration",2:(ncol(COR))]
|
|
COR_RES <- cbind(1:90,as.numeric(COR_RES)) %>% as_tibble
|
|
colnames(COR_RES) <- c("Age","Cor")
|
|
ggplot(COR_RES,aes(x=Age,y=Cor))+geom_smooth(span=0.3)+geom_point()
|
|
data.frame(COR_RES) %>% as_tibble
|
|
MIGRATION_AGE_COR <- predict(loess(Cor~Age,span=0.3,data=as.data.frame(COR_RES)))
|
|
plot(MIGRATION_AGE_COR)
|
|
|
|
#### NEXT STEPS!!!! USE CORRELATION TO DRAW FROM EACH MIGRANT IN A GIVEN YEAR
|