App combine version 1.0

Author: Annie Yang

App_combine_version/R/aqi_healthconcern.R

@@ -0,0 +1,52 @@
+#' @title Compare AQI category percentage by CBSA
+#'
+#' @description There are six categories of AQI. Each category corresponds to a different level of health concern. This function
+#'     plots percentage of each category by CBSA. It can be used to compare AQI among different CBSAs.
+#' @param data A data.frame. The default dataset is annual_aqi.
+#' @param cbsa A vector of CBSA. You can input the CBSAs you want to compare.
+#' @param category A character vector of health concern category by AQI. There are six categories.
+#'     Input category from "Good", "Moderate", "UnhealthyForSensitiveGroup", "Unhealthy", "VeryUnhealthy" and "Hazardous".
+#'     Default value includes all six categories.
+#' @param year A numeric vector.
+#' @param plot A character. The value is either "line" or "bar".
+#'     Default value is "line". You can use it to decide which kind of plot you want to get.
+#' @examples
+#' aqi_healthconcern(annual_aqi,c("Aberdeen, SD","Adrian, MI"),
+#' category =c("Good","Moderate","Unhealthy"),year=c(2000,2008,2015),plot="bar")
+#'
+#' @export
+
+aqi_healthconcern<-function(data=annual_aqi,cbsa,
+                            category=c("Good", "Moderate", "UnhealthyForSensitiveGroup", "Unhealthy", "VeryUnhealthy","Hazardous"),
+                            year=c(2000:2015),plot="line"){
+
+
+  healthconcern <- data%>%
+    mutate(
+      Good = as.numeric(data$'Good Days')/as.numeric(data$'Days with AQI'),
+      Moderate = as.numeric(data$'Moderate Days')/as.numeric(data$'Days with AQI'),
+      UnhealthyForSensitiveGroup = as.numeric(data$'Unhealthy for Sensitive Groups Days')/as.numeric(data$'Days with AQI'),
+      Unhealthy = as.numeric(data$'Unhealthy Days')/as.numeric(data$'Days with AQI'),
+      VeryUnhealthy = as.numeric(data$'Very Unhealthy Days')/as.numeric(data$'Days with AQI'),
+      Hazardous = as.numeric(data$'Hazardous Days')/as.numeric(data$'Days with AQI')
+    )%>%
+    select(CBSA,Year,category)%>%
+    filter(CBSA %in% cbsa & Year %in% year) # Calculate the percentage of each AQI category.
+
+  if(plot=="line"){
+    healthconcern %>%
+      tidyr::gather("id", "Percentage", (ncol(healthconcern)-length(category)+1):ncol(healthconcern)) %>%
+      ggplot(., aes(Year, Percentage, color=CBSA))+
+      geom_smooth(method = "loess", se=FALSE)+
+      facet_wrap(~id,scales="free_y")+
+      theme_bw()
+  } else{
+    healthconcern %>%
+      tidyr::gather("id", "Percentage", (ncol(healthconcern)-length(category)+1):ncol(healthconcern)) %>%
+      ggplot(., aes(Year, Percentage))+
+      geom_bar(stat = "identity",aes(fill = CBSA), position = "dodge")+
+      facet_wrap(~id,scales="free_y")+
+      theme_bw()
+  }
+
+}

App_combine_version/R/avg_temp.R

@@ -0,0 +1,97 @@
+#' @title Average Temperatures Trend
+#'
+#' @description This function helps you analyse global as well as countries average temperatures trend during a specified time period. 
+#'     You can use it to do comparison between countries.
+#' @param data A data.frame. Use GlobalTemperatures or GlobalLandTemperaturesByCountry dataset to do analysis.
+#' @param type A vector giving which type of trend you want to plot. 1 indicates yearly trend. 2 indicates monthly trend. c(1,2) indicates both yearly and monthly average temperature trend.
+#' @param year A vector. Time period during which you want to analyse how the temperature changed.
+#' @param month A vector. Time period during which you want to analyse how the temperature changed.
+#' @param country A vector of countries' name. Compare how the temperature changed in these countries during a specified time period.
+#' @param con Display confidence interval around smooth? (TRUE for confidence interval. FALSE by default)
+#' @examples
+#' avg_temp(data=GlobalLandTemperaturesByCountry, type=c(1,2),
+#' year=c(2000:2015), month=c(1:12), 
+#' country = c("Afghanistan","Andorra"),con="F")
+#' 
+#' @export
+
+# Load GlobalTemperatures or GlobalLandTemperaturesByCountry dataset before using this function
+
+avg_temp <- function(data, type, year, month, country, con = "FALSE"){
+  
+  if(missing(country)){ # when data=GlobalTemperatures, the country argument is missing
+    glb_avg_temp <- data %>% 
+      mutate(Month=as.numeric(format(data$dt,"%m")), # Create new column month (decimal number)
+             Month.String=format(data$dt,"%B"), # Create string month (full name)
+             Year=as.numeric(format(data$dt,"%Y")))%>% # Create new column year (4 digit)
+      select(dt, Month, Month.String, Year, LandAverageTemperature, LandAverageTemperatureUncertainty)%>%
+      na.omit() # Remove missing values
+    
+    
+    avg_temp_year<-glb_avg_temp %>% 
+      filter(Year %in% year)%>%
+      group_by(Year) %>%
+      summarise(AverageTemp=mean(LandAverageTemperature)) %>% # Calculate yearly average temperature
+      ungroup()%>%
+      ggplot(aes(x = Year, y = AverageTemp)) +
+      geom_smooth(method="loess", se=con)+
+      labs(title = "Global Yearly Average Temperatures",
+           x="Year",
+           y="Average Temperature")
+    
+    avg_temp_month<-glb_avg_temp %>%
+      filter(Year %in% year & Month %in% month)%>%
+      ggplot(aes(x=dt, y=LandAverageTemperature,colour=reorder(Month.String,-LandAverageTemperature,mean)))+
+      # Sort month from highest temperature to lowest temperature
+      geom_smooth(method="loess",se=con)+
+      labs(title="Global Average Temperatures By Month",
+           x="Year",
+           y="Average Temperature",
+           colour="Month")
+    
+    if(identical(type,c(1,2))){
+      ggpubr::ggarrange(avg_temp_year,avg_temp_month,ncol=2,nrow=1)
+    } else if(type==1) {
+      avg_temp_year
+    } else {
+      avg_temp_month
+    }
+  } else {
+    glb_avg_temp <- data %>% 
+      mutate(Month=as.numeric(format(data$dt,"%m")),
+             Month.String=format(data$dt,"%B"),
+             Year=as.numeric(format(data$dt,"%Y")))%>% # Create new columns month and year
+      select(dt, Month, Month.String, Year, Country, AverageTemperature, AverageTemperatureUncertainty)%>%
+      na.omit() %>% # Remove missing values
+      filter(Country %in% country)
+    
+    avg_temp_year <- glb_avg_temp %>% 
+      filter(Year %in% year)%>%
+      group_by(Year, Country) %>%
+      summarise(AverageTemp=mean(AverageTemperature)) %>% # Calculate yearly average temperature by country
+      ungroup()%>%
+      ggplot(aes(x = Year, y = AverageTemp,colour=Country)) +
+      geom_smooth(method="loess", se=con)+
+      labs(title = "Yearly Average Temperatures By Country",
+           x="Year",
+           y="Average Temperature")
+    
+    avg_temp_month <- glb_avg_temp %>%
+      filter(Year %in% year & Month %in% month)%>%
+      ggplot(aes(x=dt, y=AverageTemperature,colour=reorder(Month.String,-AverageTemperature,mean)))+
+      geom_smooth(method="loess",se=con)+
+      facet_wrap(~Country,scales="free")+
+      labs(title="Monthly Average Temperatures By Country",
+           x="Year",
+           y="Average Temperature",
+           colour="Month")
+    
+    if(identical(type,c(1,2))){
+      ggpubr::ggarrange(avg_temp_year,avg_temp_month)
+    } else if(type==1) {
+      avg_temp_year
+    } else {
+      avg_temp_month
+    }
+  }
+}

App_combine_version/R/boxplot_aqi.R

@@ -0,0 +1,40 @@
+#' @title Make the Boxplot of AQI
+#'
+#' @description This function helps you make boxplots of AQI in the year you are interested. And also mark the point of the CBSA you are interested.
+#'     So that you can know the AQI of the CBSA is relatively high or low compared to other geographic areas. You can make comparison of AQI in
+#'     different years by the notch. if two boxes' notches do not overlap there is ‘strong evidence’ (95% confidence) their medians differ.
+#' @param data A data.frame. The default dataset is annual_aqi.
+#' @param year A vector, years in which you want to display the distribution of AQI.
+#' @param cbsa A CBSA in US.
+#' @examples
+#' boxplot_aqi(annual_aqi,c("2000","2005"),"Albany, GA")
+#' 
+#' @export
+
+
+
+boxplot_aqi<-function(data=annual_aqi,year, cbsa){
+  
+names(data)[13] <- "MedianAQI"
+
+interested_cbsa<-data%>%filter(CBSA==cbsa & Year %in% year)
+
+box_data<-data%>%filter(Year %in% year)
+
+ggplot(box_data,aes(x=as.factor(Year), y=MedianAQI, fill=as.factor(Year)))+
+  geom_boxplot(alpha=0.2,
+               # Does the midian of AQI differ? 
+               # if two boxes' notches do not overlap there is ‘strong evidence’ (95% confidence) their medians differ.
+               notch=TRUE,
+               notchwidth = 0.5,
+               
+               outlier.colour="blue",
+               outlier.fill="blue",
+               outlier.size=2
+               )+
+  geom_point(data=interested_cbsa,
+             aes(x=as.factor(Year), y=MedianAQI),
+             color="red", size=2)+
+  labs(x= "Year")+
+  theme(legend.position="none")
+}

App_combine_version/R/install_packages.R

@@ -0,0 +1,17 @@
+
+# Install packages that the program need
+
+install_packages = function(names)
+{
+  for(name in names)
+  {
+    if (!(name %in% installed.packages()))
+      install.packages(name, repos="http://cran.us.r-project.org")
+    
+    library(name, character.only=TRUE)
+  }
+}
+
+install_packages(c("dplyr", "ggplot2",
+                   "choroplethr","choroplethrMaps","plotly","countrycode"))
+

App_combine_version/R/map_temp.R

@@ -0,0 +1,124 @@
+#' @title Plot Temperature Geographic Maps by States in USA
+#'
+#' @description This function plots temperature geographic maps for States in USA in specific year
+#' @param data A data.frame. The defalt dataset is GlobalLandTemperaturesByState.
+#' @param year numeric
+#' @examples
+#' temp_state(year=2012)
+#' 
+#' @export
+
+# Loading Packages
+# library(choroplethr)
+# library(choroplethrMaps)
+# library(plotly)
+# library(countrycode)
+
+temp_state<-function(data=GlobalLandTemperaturesByState,year){
+  
+map <- data %>%
+  mutate(Month=as.numeric(format(data$dt,"%m")), # Create new column month (decimal number)
+         Month.String=format(data$dt,"%B"), # Create string month (full name)
+         Year=as.numeric(format(data$dt,"%Y"))) %>% # Create new column year (4 digit)
+  na.omit() %>% filter(Country=="United States")
+
+map$State <- as.character(map$State)  
+map$State[map$State=="Georgia (State)"] <- "Georgia" # Changing Georgia (State)
+map$State<- as.factor(map$State)                    
+
+#' select columns of interest
+map_select <- map %>% 
+  select(Year,AverageTemperature,State) %>%
+  group_by(Year, State) %>%
+  summarise(value=mean(AverageTemperature))
+
+#Data frame must have a column named region (all lower case) and another one value.
+colnames(map_select)[2]<- "region"
+map_select$region<-tolower(map_select$region)
+
+map_state<-map_select %>%
+  filter(Year==year)
+
+map_state<-map_state[,2:3]
+
+print(state_choropleth(map_state,
+                       title = paste("Land Temperature",year," "), 
+                       num_colors = 8,
+                       legend = "Degrees"),reference_map=TRUE)
+}
+
+#' @title Plot Temperature Geographic Maps by Country
+#'
+#' @description This function plots temperature geographic maps for countries in specific year. You can use this
+#'     function to get a temperature geographic map showing the temperature change from the start year to end year.
+#' @param data A data.frame. The defalt dataset is GlobalLandTemperaturesByCountry.
+#' @param year A numeric. You can get temperature geographic maps for countries in this year.
+#' @param start A numeric. The start year you want to do temperature comparison.
+#' @param end A numeric. Then end year you want to do temperature comparison.
+#' @param diff A character. If diff=="TRUE", you will get a temperature geographic map showing the temperature
+#'     change from the start year to end year. (Default value is "FALSE")
+#' @examples
+#' temp_country(year=2012)
+#' temp_country(start=1990,end=2000,diff="TRUE")
+
+temp_country<-function(data=GlobalLandTemperaturesByCountry, year, start, end, diff="FALSE"){
+  # light grey boundaries
+  l <- list(color = toRGB("grey"), width = 0.5)
+  
+  # specify map projection/options
+  g <- list(
+    showframe = FALSE,
+    showcoastlines = FALSE,
+    projection = list(type = 'Mercator')
+  )
+ 
+  map_country <- data %>%
+    mutate(Month=as.numeric(format(data$dt,"%m")), # Create new column month (decimal number)
+           Month.String=format(data$dt,"%B"), # Create string month (full name)
+           Year=as.numeric(format(data$dt,"%Y"))) %>% # Create new column year (4 digit)
+    na.omit()%>%
+    select(Year,AverageTemperature,Country) %>%
+    group_by(Year, Country) %>%
+    summarise(AvgTemp=mean(AverageTemperature))
+  
+  code<-countrycode(map_country$Country,'country.name', 'iso3c') # Converts long country name into country codes
+  
+  map_country$CODE<-code # Create new column in map_country named "CODE"
+  
+  if(diff=="FALSE"){
+  temp<-map_country%>%filter(Year==year)
+  
+  map_temp <- plot_geo(temp) %>%
+    add_trace(
+      z = ~AvgTemp, color = ~AvgTemp, colors = 'Reds',
+      text = ~Country, locations = ~CODE, marker = list(line = l)
+    ) %>%
+    colorbar(title = 'Temperature') %>%
+    layout(
+      title = paste(year,"Temperature Map",sep=" "),
+      geo = g
+    )
+  
+  map_temp
+  } else if (diff=="TRUE") {
+    
+    temp_diff<-map_country %>% 
+      filter(Year==start | Year==end) %>% 
+      tidyr::spread(Year, AvgTemp)
+    
+    temp_diff$Difference<-unlist(temp_diff[,4]-temp_diff[,3]) # Calculate temperature variation from start year to end year
+    
+    map_temp <- plot_geo(temp_diff) %>%
+      add_trace(
+        z = ~Difference, color = ~Difference, colors = 'Reds',
+        text = ~Country, locations = ~CODE, marker = list(line = l)
+      ) %>%
+      colorbar(title = 'Temperature Variation') %>%
+      layout(
+        title = paste(start,"-",end,"Temperature Variation Map", sep=" "),
+        geo = g
+      )
+    print(map_temp)
+  }
+  
+}

App_combine_version/R/stat.R

@@ -0,0 +1,30 @@
+#' @title Calculate summary statistics for AQI
+#'
+#' @description This function helps you calculate summary statistics for AQI by CBSA.
+#' @param data A data.frame
+#' @param cbsa CBSA in US
+#' @seealso
+#' @ruturn
+#' @examples stat_func(cbsa="Ames, IA")
+#' 
+#' @export
+#' Loading packages
+#' library(dplyr)
+#' library(Rmisc)
+
+
+stat_func<-function(data=aqi,cbsa){
+  data_stat<-data%>%
+    filter(CBSA==cbsa)
+  stat_table <- as.array(summary(data_stat$`Median AQI`))
+  stat_df <- as.data.frame(stat_table)
+  colnames(stat_df) <- c("Stat", "Value")
+  Confi_Interval <- CI(data_stat$`Median AQI`, ci = 0.95) #Caculate the confidence interval of Median AQI
+  lower <- data.frame('CI lower',Confi_Interval[['lower']])
+  upper <- data.frame('CI upper',Confi_Interval[['upper']])
+  names(lower) <- c("Stat", "Value")
+  names(upper) <- c("Stat", "Value")
+  stat_df <- rbind(stat_df,lower, upper)
+  return(stat_df)
+}
+