Integrantes del equipo

Victor Manuel Muñoz Tirado - A01423434

Vivian Pérez Mosqueda - A01731984

Sebastian Romano Mena - A00831709

Instalar librerías

#install.packages("maps")
library(maps)
library(dplyr)
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
library(forecast)
## Registered S3 method overwritten by 'quantmod':
##   method            from
##   as.zoo.data.frame zoo
library(tibble)
library(ggplot2)
library(mapproj)
## Warning: package 'mapproj' was built under R version 4.3.2

Cargamos datos

US <- read.csv("historical_state_population_by_year.csv")
GA <- US[US$Estado=="GA",]
NV <- US[US$Estado=="NV",]
FL <- US[US$Estado=="FL",]
CA <- US[US$Estado=="CA",]
AL <- US[US$Estado=="AL",]

Creación de series de tiempo

ts_GA <- ts(data=GA$Población, start = c(1900,1), frequency=1)
ts_NV <- ts(data=NV$Población, start = c(1900,1), frequency=1)
ts_FL <- ts(data=FL$Población, start = c(1900,1), frequency=1)
ts_CA <- ts(data=CA$Población, start = c(1900,1), frequency=1)
ts_AL <- ts(data=AL$Población, start = c(1900,1), frequency=1)

Creación de modelos

arima_GA <- auto.arima(ts_GA)
arima_NV <- auto.arima(ts_NV)
arima_FL <- auto.arima(ts_FL)
arima_CA <- auto.arima(ts_CA)
arima_AL <- auto.arima(ts_AL)

Generamos pronósticos

pronostico_GA <- forecast(arima_GA, level=c(95), h=51)
pronostico_NV <- forecast(arima_NV, level=c(95), h=51)
pronostico_FL <- forecast(arima_FL, level=c(95), h=51)
pronostico_CA <- forecast(arima_CA, level=c(95), h=51)
pronostico_AL <- forecast(arima_AL, level=c(95), h=51)

Gráficos de pronósticos

plot(pronostico_GA, main= "Población GA")

plot(pronostico_NV, main= "Población NV")

plot(pronostico_FL, main= "Población FL")

plot(pronostico_CA, main= "Población CA")

plot(pronostico_AL, main= "Población AL")

Generamos mapas con la información

map_GA <- as.data.frame(pronostico_GA)
map_NV <- as.data.frame(pronostico_NV)
map_FL <- as.data.frame(pronostico_FL)
map_CA <- as.data.frame(pronostico_CA)
map_AL <- as.data.frame(pronostico_AL)
map_GA <- map_GA %>% rownames_to_column("Años")
map_NV <- map_NV %>% rownames_to_column("Años")
map_FL <- map_FL %>% rownames_to_column("Años")
map_CA <- map_CA %>% rownames_to_column("Años")
map_AL <- map_AL %>% rownames_to_column("Años")
map_GA$region <- "georgia"
map_NV$region <- "nevada"
map_FL$region <- "florida"
map_CA$region <- "california"
map_AL$region <- "alabama"
map_GA_30 <- filter(map_GA, Años == 2030)
map_NV_30 <- filter(map_NV, Años == 2030)
map_FL_30 <- filter(map_FL, Años == 2030)
map_CA_30 <- filter(map_CA, Años == 2030)
map_AL_30 <- filter(map_AL, Años == 2030)

map_2030 <- rbind(map_GA_30,map_NV_30,map_FL_30,map_CA_30,map_AL_30)

map_GA_40 <- filter(map_GA, Años == 2040)
map_NV_40 <- filter(map_NV, Años == 2040)
map_FL_40 <- filter(map_FL, Años == 2040)
map_CA_40 <- filter(map_CA, Años == 2040)
map_AL_40 <- filter(map_AL, Años == 2040)

map_2040 <- rbind(map_GA_40,map_NV_40,map_FL_40,map_CA_40,map_AL_40)

map_GA_50 <- filter(map_GA, Años == 2050)
map_NV_50 <- filter(map_NV, Años == 2050)
map_FL_50 <- filter(map_FL, Años == 2050)
map_CA_50 <- filter(map_CA, Años == 2050)
map_AL_50 <- filter(map_AL, Años == 2050)

map_2050 <- rbind(map_GA_50,map_NV_50,map_FL_50,map_CA_50,map_AL_50)

map_GA_60 <- filter(map_GA, Años == 2060)
map_NV_60 <- filter(map_NV, Años == 2060)
map_FL_60 <- filter(map_FL, Años == 2060)
map_CA_60 <- filter(map_CA, Años == 2060)
map_AL_60 <- filter(map_AL, Años == 2060)

map_2060 <- rbind(map_GA_60,map_NV_60,map_FL_60,map_CA_60,map_AL_60)

map_GA_70 <- filter(map_GA, Años == 2070)
map_NV_70 <- filter(map_NV, Años == 2070)
map_FL_70 <- filter(map_FL, Años == 2070)
map_CA_70 <- filter(map_CA, Años == 2070)
map_AL_70 <- filter(map_AL, Años == 2070)

map_2070 <- rbind(map_GA_70,map_NV_70,map_FL_70,map_CA_70,map_AL_70)
map_2030$Poblacion <- map_2030$`Point Forecast`
map_2040$Poblacion <- map_2040$`Point Forecast`
map_2050$Poblacion <- map_2050$`Point Forecast`
map_2060$Poblacion <- map_2060$`Point Forecast`
map_2070$Poblacion <- map_2070$`Point Forecast`
map_usa <- map_data("state")
map_usa_2030 <- left_join(map_usa, map_2030, by = "region")
map_usa_2040 <- left_join(map_usa, map_2040, by = "region")
map_usa_2050 <- left_join(map_usa, map_2050, by = "region")
map_usa_2060 <- left_join(map_usa, map_2060, by = "region")
map_usa_2070 <- left_join(map_usa, map_2070, by = "region")
ggplot(data = map_usa_2030) + 
  geom_polygon(aes(x = long, y = lat, group = group, fill = Poblacion), color = "black") +
  scale_fill_gradient(low = "lightblue", high = "darkblue", labels = scales::comma) +
  coord_map() +
  labs(title="Población 2030 (en millones)", fill = "Población\n(en millones)")

ggplot(data = map_usa_2040) + 
  geom_polygon(aes(x = long, y = lat, group = group, fill =Poblacion), color = "black") +
  scale_fill_gradient(low = "lightblue", high = "darkblue", labels = scales::comma) +
  coord_map() +
  labs(title="Población 2040 (en millones)", fill = "Población\n(en millones)")

ggplot(data = map_usa_2050) + 
  geom_polygon(aes(x = long, y = lat, group = group, fill =Poblacion), color = "black") +
  scale_fill_gradient(low = "lightblue", high = "darkblue", labels = scales::comma) +
  coord_map() +
  labs(title="Población 2050 (en millones)", fill = "Población\n(en millones)")

ggplot(data = map_usa_2060) + 
  geom_polygon(aes(x = long, y = lat, group = group, fill =Poblacion), color = "black") +
  scale_fill_gradient(low = "lightblue", high = "darkblue", labels = scales::comma) +
  coord_map() +
  labs(title="Población 2060 (en millones)", fill = "Población\n(en millones)")

ggplot(data = map_usa_2070) + 
  geom_polygon(aes(x = long, y = lat, group = group, fill =Poblacion), color = "black") +
  scale_fill_gradient(low = "lightblue", high = "darkblue", labels = scales::comma) +
  coord_map() +
  labs(title="Población 2070 (en millones)", fill = "Población\n(en millones)")