library(tidyverse)
LIN_1999 <- read_csv("Data/Raw_Data/Mortality_Rates_New/Lincoln_Age_Adjusted_1999-2020.csv") %>% select(Year,Sex,Mort_Rate=`Age Adjusted Rate`)%>% mutate(Region='Lincoln') 
WY_1999 <- read_csv("Data/Raw_Data/Mortality_Rates_New/Wyoming_Age_Adjusted_1999-2020.csv") %>% select(Year,Sex,Mort_Rate=`Age Adjusted Rate`)%>% mutate(Region='Wyoming') %>% filter(Year<2018)
US_1999 <- read_csv("Data/Raw_Data/Mortality_Rates_New/US_Age_Adjusted_1999-2020.csv")%>% select(Year,Sex,Mort_Rate=`Age Adjusted Rate`) %>% filter(!is.na(Sex),!is.na(Year)) %>% mutate(Region="US")%>% filter(Year<2018)


WY_2018 <- read_csv("Data/Raw_Data/Mortality_Rates_New/Wyoming_Age_Adjusted_2018-2023.csv") %>% select(Year,Sex,Mort_Rate=`Age Adjusted Rate`) %>% mutate(Region='Wyoming') 
US_2018 <- read_csv("Data/Raw_Data/Mortality_Rates_New/US_Age_Adjusted_2018-2023.csv")%>% select(Year,Sex,Mort_Rate=`Age Adjusted Rate`)  %>% mutate(Region="US")
##No adjustment for later data allowed
LIN_2018<- read_csv("Data/Raw_Data/Mortality_Rates_New/Lincoln_Not_Age_Adjusted_2018-2023.csv") %>% select(Year,Sex,Mort_Rate=`Crude Rate`)%>% mutate(Region='Lincoln') 
ADJUST_TERM <- LIN_2018 %>% rename(UNADJUSTED=Mort_Rate) %>% inner_join(LIN_1999) %>% filter(!is.na(Year)) %>% mutate(Ratio=Mort_Rate/UNADJUSTED) %>% group_by(Sex) %>% summarize(Ratio=mean(Ratio))
LIN_2018 <- LIN_2018 %>% filter(Year>2020) %>% left_join(ADJUST_TERM) %>% mutate(Mort_Rate=Mort_Rate*Ratio) %>% select(-Ratio)

DF <- rbind(LIN_1999,LIN_2018,WY_1999,US_1999,US_2018,WY_2018) %>% filter(!is.na(Year),!is.na(Sex))
ggplot(DF,aes(x=Year,y=Mort_Rate,group=Region,color=Region,fill=Region))+geom_point()+geom_smooth(method="lm")+ facet_grid(. ~ Sex)
ggplot(DF,aes(x=Year,y=Mort_Rate,group=Region,color=Region,fill=Region))+geom_point()+geom_smooth()+ facet_grid(. ~ Sex)
ggplot(DF,aes(x=Year,y=Mort_Rate,group=Region,color=Region,fill=Region))+geom_point()+ facet_grid(. ~ Sex)
ggplot(DF,aes(x=Year,y=Mort_Rate,group=Region,color=Region,fill=Region))+geom_point()+geom_smooth()


REG_DATA <- DF %>% pivot_wider(values_from=Mort_Rate,names_from=Region)
library(fixest)
REG_DATA
feols(log(Lincoln)~log(US)+Sex+Year,REG_DATA)
feols((Lincoln)~Sex+Year,REG_DATA)




%>% select(Year,Sex,County,Mort_Rate=`Age Adjusted Rate`) %>% filter(!is.na(County))