# New Daily US Covid Cases & Deaths
# Bob Agnew, raagnew1@gmail.com, raagnew.com 

library(graphics)
end_date <- "Jul 23 2021" # Matched to end date on New York Times datasets
options(scipen=999)

# TOTAL US
popn <- 328239523 # July 1 2019 per Census
x <- read.csv("c:/COVID/US.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us.csv raw data)
total_cases <- as.vector(x[-(1:37),2])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 59)
tc <- ts(tc, frequency = 1, start = 59)
total_cases <- ts(total_cases, frequency = 1, start = 58)
pdf("c:/COVID/US_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily US Covid Cases from Feb 28 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=","),"                    Ending Daily Trend = ",formatC(round(tail(tc,1)),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total US Covid Cases % of Population from Feb 27 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:44),3])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 66)
tc <- ts(tc, frequency = 1, start = 66)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 65 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 65 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 65 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 65 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily US Covid Deaths from Mar 6 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=","),"                    Ending Daily Trend = ",formatC(round(tail(tc,1)),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling US Covid Deaths % of Lagged Rolling US Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# SELECTED STATES

# ARIZONA
popn <- 7278717 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Arizona",]
total_cases <- as.vector(x[-(1:47),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 74)
tc <- ts(tc, frequency = 1, start = 74)
total_cases <- ts(total_cases, frequency = 1, start = 73)
pdf("c:/COVID/AZ_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Arizona Covid Cases from Mar 14 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Arizona Covid Cases % of Population from Mar 13 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:59),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 86)
tc <- ts(tc, frequency = 1, start = 86)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 85 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 85 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 85 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 85 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Arizona Covid Deaths from Mar 26 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Arizona Covid Deaths % of Lagged Rolling Arizona Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# CALIFORNIA
popn <- 39512223 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="California",]
total_cases <- as.vector(x[-(1:33),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 59)
tc <- ts(tc, frequency = 1, start = 59)
total_cases <- ts(total_cases, frequency = 1, start = 58)
pdf("c:/COVID/CA_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily California Covid Cases from Feb 28 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total California Covid Cases % of Population from Feb 27 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:49),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 75)
tc <- ts(tc, frequency = 1, start = 75)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 74 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 74 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 74 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 74 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily California Covid Deaths from Mar 15 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling California Covid Deaths % of Lagged Rolling California Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# FLORIDA
popn <- 21477737 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Florida",]
total_cases <- as.vector(x[-(1:13),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 75)
tc <- ts(tc, frequency = 1, start = 75)
total_cases <- ts(total_cases, frequency = 1, start = 74)
pdf("c:/COVID/FL_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Florida Covid Cases from Mar 15 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Florida Covid Cases % of Population from Mar 14 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:14),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 76)
tc <- ts(tc, frequency = 1, start = 76)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 75 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 75 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 75 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 75 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Florida Covid Deaths from Mar 16 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Florida Covid Deaths % of Lagged Rolling Florida Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# ILLINOIS
popn <- 12671821 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Illinois",]
total_cases <- as.vector(x[-(1:43),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 68)
tc <- ts(tc, frequency = 1, start = 68)
total_cases <- ts(total_cases, frequency = 1, start = 67)
pdf("c:/COVID/IL_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Illinois Covid Cases from Mar 8 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Illinois Covid Cases % of Population from Mar 7 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:54),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 79)
tc <- ts(tc, frequency = 1, start = 79)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 78 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 78 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 78 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 78 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Illinois Covid Deaths from Mar 19 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Illinois Covid Deaths % of Lagged Rolling Illinois Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# MICHIGAN
popn <- 9986857 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Michigan",]
total_cases <- as.vector(x[-(1:1),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 72)
tc <- ts(tc, frequency = 1, start = 72)
total_cases <- ts(total_cases, frequency = 1, start = 71)
pdf("c:/COVID/MI_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Michigan Covid Cases from Mar 12 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Michigan Covid Cases % of Population from Mar 11 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:9),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 80)
tc <- ts(tc, frequency = 1, start = 80)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 79 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 79 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 79 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 79 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Michigan Covid Deaths from Mar 20 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Michigan Covid Deaths % of Lagged Rolling Michigan Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# NEW YORK
popn <- 19453561 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="New York",]
total_cases <- as.vector(x[-(1:1),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 63)
tc <- ts(tc, frequency = 1, start = 63)
total_cases <- ts(total_cases, frequency = 1, start = 62)
pdf("c:/COVID/NY_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily New York Covid Cases from Mar 3 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("top",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total New York Covid Cases % of Population from Mar 2 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:13),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 75)
tc <- ts(tc, frequency = 1, start = 75)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 74 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 74 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 74 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 74 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily New York Covid Deaths from Mar 15 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("top",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling New York Covid Deaths % of Lagged Rolling New York Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# NORTH CAROLINA
popn <- 10488084 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="North Carolina",]
total_cases <- as.vector(x[-(1:7),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 71)
tc <- ts(tc, frequency = 1, start = 71)
total_cases <- ts(total_cases, frequency = 1, start = 70)
pdf("c:/COVID/NC_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily North Carolina Covid Cases from Mar 11 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total North Carolina Covid Cases % of Population from Mar 10 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:27),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 91)
tc <- ts(tc, frequency = 1, start = 91)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 90 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 90 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 90 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 90 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily North Carolina Covid Deaths from Mar 31 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling North Carolina Covid Deaths % of Lagged Rolling North Carolina Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# TEXAS
popn <- 28995881 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Texas",]
total_cases <- as.vector(x[-(1:24),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 68)
tc <- ts(tc, frequency = 1, start = 68)
total_cases <- ts(total_cases, frequency = 1, start = 67)
pdf("c:/COVID/TX_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Texas Covid Cases from Mar 8 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Texas Covid Cases % of Population from Mar 7 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:40),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 84)
tc <- ts(tc, frequency = 1, start = 84)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 83 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 83 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 83 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 83 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Texas Covid Deaths from Mar 24 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Texas Covid Deaths % of Lagged Rolling Texas Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# PENNSYLVANIA
popn <- 12801989 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Pennsylvania",]
total_cases <- as.vector(x[,4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 67)
tc <- ts(tc, frequency = 1, start = 67)
total_cases <- ts(total_cases, frequency = 1, start = 66)
pdf("c:/COVID/PA_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Pennsylvania Covid Cases from Mar 7 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Pennsylvania Covid Cases % of Population from Mar 6 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:14),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 81)
tc <- ts(tc, frequency = 1, start = 81)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 80 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 80 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 80 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 80 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Pennsylvania Covid Deaths from Mar 21 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Pennsylvania Covid Deaths % of Lagged Rolling Pennsylvania Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# GEORGIA
popn <- 10617423 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Georgia",]
total_cases <- as.vector(x[-(1:10),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 73)
tc <- ts(tc, frequency = 1, start = 73)
total_cases <- ts(total_cases, frequency = 1, start = 72)
pdf("c:/COVID/GA_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Georgia Covid Cases from Mar 13 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Georgia Covid Cases % of Population from Mar 12 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:21),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 84)
tc <- ts(tc, frequency = 1, start = 84)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 83 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 83 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 83 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 83 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Georgia Covid Deaths from Mar 24 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Georgia Covid Deaths % of Lagged Rolling Georgia Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# OHIO
popn <- 11689100 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Ohio",]
total_cases <- as.vector(x[-(1:1),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 71)
tc <- ts(tc, frequency = 1, start = 71)
total_cases <- ts(total_cases, frequency = 1, start = 70)
pdf("c:/COVID/OH_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Ohio Covid Cases from Mar 11 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Ohio Covid Cases % of Population from Mar 10 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:13),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 83)
tc <- ts(tc, frequency = 1, start = 83)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 82 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 82 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 82 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 82 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Ohio Covid Deaths from Mar 23 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Ohio Covid Deaths % of Lagged Rolling Ohio Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# WISCONSIN
popn <- 5822434 # July 1 2019 per Census
x <- read.csv("c:/COVID/STATES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-states.csv raw data)
x <- x[trimws(x[,2])=="Wisconsin",]
total_cases <- as.vector(x[-(1:32),4])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 69)
tc <- ts(tc, frequency = 1, start = 69)
total_cases <- ts(total_cases, frequency = 1, start = 68)
pdf("c:/COVID/WI_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Wisconsin Covid Cases from Mar 9 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Wisconsin Covid Cases % of Population from Mar 8 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:53),5])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 90)
tc <- ts(tc, frequency = 1, start = 90)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 89 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 89 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 89 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 89 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Wisconsin Covid Deaths from Mar 30 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Wisconsin Covid Deaths % of Lagged Rolling Wisconsin Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# SELECTED COUNTIES

# COOK, ILLINOIS
popn <- 5150233 # July 1 2019 per Census
x <- read.csv("c:/COVID/COUNTIES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-counties.csv raw data)
x <- x[trimws(x[,2])=="Cook"&trimws(x[,3])=="Illinois",]
total_cases <- as.vector(x[-(1:52),5])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 77)
tc <- ts(tc, frequency = 1, start = 77)
total_cases <- ts(total_cases, frequency = 1, start = 76)
pdf("c:/COVID/COOK_IL_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Cook Cty IL Covid Cases from Mar 17 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Cook Cty IL Covid Cases % of Population from Mar 16 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:54),6])
k <- length(total_deaths)
daily_deaths <- pmax(1,total_deaths[-1]-total_deaths[-k])
s <- ts(daily_deaths, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 79)
tc <- ts(tc, frequency = 1, start = 79)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 78 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 78 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 78 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 78 + k - 172)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Deaths",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Cook Cty IL Covid Deaths from Mar 19 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Cook Cty IL Covid Deaths % of Lagged Rolling Cook Cty IL Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# BEXAR, TEXAS
popn <- 2003554 # July 1 2019 per Census
x <- read.csv("c:/COVID/COUNTIES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-counties.csv raw data)
x <- x[trimws(x[,2])=="Bexar"&trimws(x[,3])=="Texas",]
total_cases <- as.vector(x[-(1:40),5])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 84)
tc <- ts(tc, frequency = 1, start = 84)
total_cases <- ts(total_cases, frequency = 1, start = 83)
pdf("c:/COVID/BEXAR_TX_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Bexar Cty TX Covid Cases from Mar 24 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Bexar Cty TX Covid Cases % of Population from Mar 23 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:40),6])
k <- length(total_deaths)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 83 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 83 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 83 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 83 + k - 172)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Bexar Cty TX Covid Deaths % of Lagged Rolling Bexar Cty TX Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# LAKE, ILLINOIS
popn <- 696535 # July 1 2019 per Census
x <- read.csv("c:/COVID/COUNTIES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-counties.csv raw data)
x <- x[trimws(x[,2])=="Lake"&trimws(x[,3])=="Illinois",]
total_cases <- as.vector(x[-(1:1),5])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 73)
tc <- ts(tc, frequency = 1, start = 73)
total_cases <- ts(total_cases, frequency = 1, start = 72)
pdf("c:/COVID/LAKE_IL_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Lake Cty IL Covid Cases from Mar 13 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Lake Cty IL Covid Cases % of Population from Mar 12 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:17),6])
k <- length(total_deaths)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 88 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 88 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 88 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 88 + k - 172)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Lake Cty IL Covid Deaths % of Lagged Rolling Lake Cty IL Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# FLAGLER, FLORIDA
popn <- 115081 # July 1 2019 per Census
x <- read.csv("c:/COVID/COUNTIES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-counties.csv raw data)
x <- x[trimws(x[,2])=="Flagler"&trimws(x[,3])=="Florida",]
total_cases <- as.vector(x[-(1:7),5])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 89)
tc <- ts(tc, frequency = 1, start = 89)
total_cases <- ts(total_cases, frequency = 1, start = 88)
pdf("c:/COVID/FLAGLER_FL_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Flagler Cty FL Covid Cases from Mar 29 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Flagler Cty FL Covid Cases % of Population from Mar 28 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:12),6])
k <- length(total_deaths)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 94 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 94 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 94 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 94 + k - 172)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Flagler Cty FL Covid Deaths % of Lagged Rolling Flagler Cty FL Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# CHEMUNG, NEW YORK
popn <- 83456 # July 1 2019 per Census
x <- read.csv("c:/COVID/COUNTIES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-counties.csv raw data)
x <- x[trimws(x[,2])=="Chemung"&trimws(x[,3])=="New York",]
total_cases <- as.vector(x[-(1:10),5])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 95)
tc <- ts(tc, frequency = 1, start = 95)
total_cases <- ts(total_cases, frequency = 1, start = 94)
pdf("c:/COVID/CHEMUNG_NY_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Chemung Cty NY Covid Cases from Apr 4 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Chemung Cty NY Covid Cases % of Population from Apr 3 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:16),6])
k <- length(total_deaths)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 100 + k - 200)
roll_deaths <- ts(pmax(0,total_deaths[-(1:28)] - total_deaths[-(173:200)]),frequency = 1, start = 100 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 100 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 100 + k - 172)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Chemung Cty NY Covid Deaths % of Lagged Rolling Chemung Cty NY Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()

# CUMBERLAND, PENNSYLVANIA
popn <- 253370 # July 1 2019 per Census
x <- read.csv("c:/COVID/COUNTIES.csv",header=TRUE) # From github.com/nytimes/covid-19-data (us-counties.csv raw data)
x <- x[trimws(x[,2])=="Cumberland"&trimws(x[,3])=="Pennsylvania",]
total_cases <- as.vector(x[-(1:12),5])
k <- length(total_cases)
daily_cases <- pmax(1,total_cases[-1]-total_cases[-k])
s <- ts(daily_cases, frequency = 7, start = 1)
logs <- ts(log(s), frequency = 7, start = 1)
t <- stl(logs,"per")
tc <- exp(t$time.series[,2])
s <- ts(s, frequency = 1, start = 86)
tc <- ts(tc, frequency = 1, start = 86)
total_cases <- ts(total_cases, frequency = 1, start = 85)
pdf("c:/COVID/CUMBERLAND_PA_COVID_CASES_AND_DEATHS.pdf",width=10,height=7,onefile=TRUE)
ts.plot(s,tc,gpars=list(lty=rep(1,2),lwd=rep(2,2),ylab="New Daily Covid Cases",xlab="Days since 2019",tck=1,col=c("blue","red"),main=paste0("New Daily Cumberland Cty PA Covid Cases from Mar 26 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
legend("topleft",bty="n",lty=rep(1,2),lwd=rep(2,2),col=c("blue","red"),legend=c("Actual","Trend"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
ts.plot(100*total_cases/popn,gpars=list(lty=1,lwd=2,ylab="Total Covid Cases % of Population",xlab="Days since 2019",tck=1,col="blue",main=paste0("Total Cumberland Cty PA Covid Cases % of Population from Mar 25 2020 to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Cases = ",formatC(tail(total_cases,1),format="d",big.mark=",")),side=3)
total_deaths <- as.vector(x[-(1:44),6])
k <- length(total_deaths)
total_deaths <- ts(tail(total_deaths,200), frequency = 1, start = 117 + k - 200)
roll_deaths <- ts(total_deaths[-(1:28)] - total_deaths[-(173:200)],frequency = 1, start = 117 + k - 172)
total_cases <- ts(tail(total_cases,221)[-(201:221)], frequency = 1, start = 117 + k - 200) # Cases lagged by 21 days
roll_cases <- ts(total_cases[-(1:28)] - total_cases[-(173:200)],frequency = 1, start = 117 + k - 172)
ts.plot(100*roll_deaths/roll_cases,gpars=list(lty=1,lwd=2,ylab="28-Day Rolling Covid Deaths % of Rolling Covid Cases lagged 21 Days",xlab="Days since 2019",tck=1,col="blue",main=paste0("Rolling Cumberland Cty PA Covid Deaths % of Lagged Rolling Cumberland Cty PA Covid Cases to ",end_date),sub="Data from New York Times : github.com/nytimes/covid-19-data"))
mtext(paste0("Total Deaths = ",formatC(tail(total_deaths,1),format="d",big.mark=",")),side=3)
dev.off()