sormehy_final.Rmd

---
title: "NES Industry Analysis in 4 States"
resource_files:
- cb_2018_us_state_500k.shx
- cb_2018_us_state_500k.shp.iso.xml
- cb_2018_us_state_500k.shp.ea.iso.xml
- cb_2018_us_state_500k.prj
- cb_2018_us_state_500k.dbf
- cb_2018_us_state_500k.cpg
runtime: shiny
output:
  flexdashboard::flex_dashboard:
    orientation: rows
    source_code: embed
    theme: flatly
---

```{r setup, include=FALSE}
#knitr::opts_knit$set(root.dir = normalizePath())
knitr::opts_chunk$set(echo = FALSE)
```


```{r context="setup", include=FALSE}
library(shiny)
library(ggplot2)
library(dplyr)
library(tidyr)
library(splines)
library(stringr)
library(RColorBrewer)
library(reshape2)
library(plotly)
library(leaflet)
library(DT)
require(rgdal)
require(leaflet.extras)
require(readxl)

```


```{r data}

## load raw data
#data <- read.csv("nes_selectdata.csv")
data <- read.csv("nes_extratrim.csv")
#data <- read.csv("~/sormehy_final/nes_trimmed.csv")

## Make the data easier to read for the user
nes_st0418_p <- data %>% mutate(Industry.Title = ifelse(NAICS == 23, 'Construction',
                  ifelse(NAICS == 4853, 'Taxi & Limousine Service', ifelse(NAICS == 484, 'Truck Transportation', ifelse(NAICS == 4841, 'General Freight Trucking',
                  ifelse(NAICS == 48411, 'General Freight Trucking, Local', ifelse(NAICS == 48412, 'General Freight Trucking, Long-Distance', 
                  ifelse(NAICS == 4842, 'Specialized Freight Trucking', ifelse(NAICS == 236, 'Construction of Buildings', 
                  ifelse(NAICS == 2361, 'Residential Building Construction', ifelse(NAICS == 2362, 'Nonresidential Building Construction', 
                  ifelse(NAICS == 237, 'Heavy & Civil Engineering Construction', ifelse(NAICS == 2371, 'Utility System Construction',
                  ifelse(NAICS == 2372 | 23721, 'Land Subdivision', ifelse(NAICS == 2373 | 23731, 'Highway, Street, & Bridge Construction', 
                  ifelse(NAICS == 2379 | 23799, 'Other Heavy & Civ Eng Construction', ifelse(NAICS == 238, "Specialty Trade Contractors",
                    ifelse(NAICS == 2381, "Foundation, Structure, & Building Exterior Contractors",
                    ifelse(NAICS == 2382, "Building Equipment Contractors",
                    ifelse(NAICS == 23821, "Electrical Contractors and Other Wiring Installation Contractors",
                    ifelse(NAICS == 23822, "Plumbing, Heating, and Air-Conditioning Contractors",
                    ifelse(NAICS == 23829, "Other Building Equipment Contractors",
                    ifelse(NAICS == 2383, "Building Finishing Contractors",
                    ifelse(NAICS == 2389, "Other Specialty Trade Contractors"))))))))))))))))))))))))


#construction <- pa.ind.emp.raw %>%
#  filter(str_detect(NAICS, "^23"))


NAICS.Categories = c("^4853", "^23", "^48")
Big.Industries = c('Taxi & Limousine Service', 'Construction', 'Truck Transportation')
Count.Type = c("Total_RCPTOT", "Total_ESTAB")
Ind.Industries = c("Taxi & Limousine Service", "Truck Transportation", "General Freight Trucking", "General Freight Trucking, Local", 
           "General Freight Trucking, Long-Distance", "Specialized Freight Trucking", "Construction", "Construction of Buildings", 
           "Residential Building Construction", "Nonresidential Building Construction", "Heavy & Civil Engineering Construction",
           "Utility System Construction", "Land Subdivision", "Highway, Street, & Bridge Construction",
           "Other Heavy & Civ Eng Construction",
           "Specialty Trade Contractors", "Foundation, Structure, & Building Exterior Contractors", "Building Equipment Contractors", 
           "Electrical Contractors and Other Wiring Installation Contractors", "Plumbing, Heating, and Air-Conditioning Contractors",
           "Other Building Equipment Contractors", "Building Finishing Contractors", "Other Specialty Trade Contractors")
State.op <- c("Pennsylvania", "Ohio", "Illinois", "Montana")
Year.op <- c(2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017, 2018)

## PICK THE THREE out of FOUR STATES FOR COMPARISON
nes_st0418_p <- nes_st0418_p %>% mutate(ST_n = ifelse(ST == 17, 'Illinois', ifelse(ST == 39, 'Ohio', ifelse(ST == 42, 'Pennsylvania', ifelse(ST == 30, 'Montana', 'NaN')))))

#print(getwd())

```

Inputs Column {.sidebar}
-----------------------------------------------------------------------

```{r context="render"}


selectInput("naics_var", label = "Select NAICS Category:", choices = NAICS.Categories)

selectInput("big_ind", label = "Select Industry Title Categories:", choices = Big.Industries)

selectInput("count_type", label = "Select Type of Count:", choices = Count.Type)


checkboxGroupInput("ch_state", "Select States to Compare:", choices = State.op,
                  selected = c("Pennsylvania", "Ohio", "Illinois"), inline = TRUE
                  )

# checkboxGroupInput("ind_ind", "Select Detailed Industries:", choices = Ind.Industries,
#                   selected = c("Taxi and Limousine Service", "Truck Transportation", "Construction"), inline = TRUE
#                   )
selectizeInput("ind_ind", "Select Detailed Industries:", choices= Ind.Industries, 
               selected = c("Taxi and Limousine Service", "Truck Transportation", "Construction"),
               multiple = TRUE)

selectizeInput("yr", "Select Year:", choices= Year.op, 
               selected = c("2012"),
               multiple = FALSE)

downloadButton("downloadFile", label = "Download Data")
downloadHandler(filename = function() {
     paste('resultset-', Sys.Date(), '.csv', sep='')
   },
     content = function(file) {
     write.csv(industry_subset(), file, row.names = FALSE)
   }
)

```


Page with Tabset {.tabset .tabset-fade}
-----------------------------------------------------------------------

### Time Series Comparing States across Industries

```{r context="server"}

industry_subset <- reactive({
    nes_st0418_p[nes_st0418_p$Industry.Title %in% input$ind_ind & nes_st0418_p$ST_n %in% input$ch_state,]
})


data_year <- reactive({
    filter(data, 
      input$big_ind
    )
})

count_type <- reactive({
    filter(nes_st0418_p, input$count_type)
})

output$mod.plot <- renderPlotly({
  
  if(input$employed_year == "Emp.2018"){
    ggplotly(ggplot(data = industry_subset(), mapping = aes(x = reorder(Industry.Title, -Emp.2018), xend = reorder(Industry.Title, -Emp.2018), y = Emp.2018, yend = 0, color = Industry.Title)) + 
    geom_point() +
      geom_segment()+
    coord_flip() + 
    labs(x='', y='Number of Employees', title='Pennsylvanians Employed in Given Industry') + theme(legend.position = "none"))
  }
  else {
    ggplotly(ggplot(data = industry_subset(), mapping = aes(x = reorder(Industry.Title, -Emp.2028), xend = reorder(Industry.Title, -Emp.2028), y = Emp.2028, yend = 0, color = Industry.Title)) + 
    geom_point() +
      geom_segment()+
    coord_flip() + 
    labs(x='', y='Number of Employees', title='Pennsylvanians Employed in Given Industry') + theme(legend.position = "none"))
  }
    
})

output$compare.plot <- renderPlotly({
    if(input$count_type == 'Total_RCPTOT'){
  ggplotly(ggplot(data = industry_subset(), mapping = aes(x=Year, y=(Total_RCPTOT/1000000), color = ST_n)) +
    geom_line() + geom_point() +
  labs(title='Industry Growth Over Time (2004 - 2018)', 
       x = 'Year', y = 'Total Receipts (in millions) in Given State', color="State") +
    facet_wrap('Industry.Title') + theme(
plot.title = element_text(size=12, face="bold"),
axis.title.x = element_text(size=13),
axis.title.y = element_text(size=13)))
    }else{
        ggplotly(ggplot(data = industry_subset(), mapping = aes(x=Year, y=(Total_ESTAB/10000), color = ST_n)) +
    geom_line() + geom_point() +
  labs(title='Industry Growth Over Time (2004 - 2018)', 
       x = 'Year', y = 'Total Establishments (in 10,000s)', color="State") +
    facet_wrap('Industry.Title') + theme(
plot.title = element_text(size=12, face="bold"),
axis.title.x = element_text(size=13),
axis.title.y = element_text(size=13)))
    }
})


```

```{r context="render"}
#plotlyOutput("scatter.ordered")
#plotlyOutput("mod.plot")
plotlyOutput("compare.plot")
```


### Time Series Comparing Industries across States

```{r context="server"}

industry_subset <- reactive({
    nes_st0418_p[nes_st0418_p$Industry.Title %in% input$ind_ind & nes_st0418_p$ST_n %in% input$ch_state,]
})


data_year <- reactive({
    filter(data, 
      input$big_ind
    )
})

count_type <- reactive({
    filter(nes_st0418_p, input$count_type)
})



output$compare.plot2 <- renderPlotly({
    if(input$count_type == 'Total_RCPTOT'){
  ggplotly(ggplot(data = industry_subset(), mapping = aes(x=Year, y=(Total_RCPTOT/1000000), color = Industry.Title)) +
    geom_line() + geom_point() +
  labs(title='Industry Growth Over Time (2004 - 2018)', 
       x = 'Year', y = 'Total Receipts (in millions) in Given Industry', color="Industry") +
    facet_wrap('ST_n') + theme(
plot.title = element_text(size=12, face="bold"),
axis.title.x = element_text(size=13),
axis.title.y = element_text(size=13)))
    }else{
        ggplotly(ggplot(data = industry_subset(), mapping = aes(x=Year, y=(Total_ESTAB/10000), color = Industry.Title)) +
    geom_line() + geom_point() +
  labs(title='Industry Growth Over Time (2004 - 2018)', 
       x = 'Year', y = 'Total Establishments (in 10,000s)', color="Industry") +
    facet_wrap('ST_n') + theme(
plot.title = element_text(size=12, face="bold"),
axis.title.x = element_text(size=13),
axis.title.y = element_text(size=13)))
    }
})


```

```{r context="render"}

plotlyOutput("compare.plot2")
```


### Mapping

```{r, controls}
# leaflet() %>%
#   # Basemaps
#   addTiles(group = "OSM (default)") %>%
#   addProviderTiles("Stamen.Toner", group = "Toner") %>%
#   addProviderTiles("Stamen.TonerLite", group = "Toner Lite") %>%
#   # Layers control
#   addLayersControl(
#     baseGroups = c("OSM (default)", "Toner", "Toner Lite"),
#     options = layersControlOptions(collapsed = FALSE)
#   )
```

```{r, polygon}
# op_data <- read_excel("Geographies_of_Opportunity_Ranking_Well_Being_by_Congressional_District_(114th_Congress).xlsx") %>%
  # mutate(Number = str_pad(Number, 4, pad = "0"))
# library(raster)
# cds.load <- shapefile("cb_2018_us_state_500k.shp")

cds.load <- readOGR("cb_2018_us_state_500k.shp", layer = "cb_2018_us_state_500k", GDAL1_integer64_policy = TRUE)

# Just having the matching GEOID's
cds <- cds.load[cds.load$GEOID %in% nes_st0418_p$ST ,]

cds@data <- merge(cds@data, nes_st0418_p, sort = FALSE, by.x = "GEOID", by.y = "ST")

#cds <- reactive({cds[cds$Industry.Title %in% input$big_ind & cds$Year %in% input$yr,]})

pal <- colorNumeric(
  palette = "viridis",
  domain = cds$`Total_RCPTOT`)
pal2 <- colorNumeric(
  palette = "Blues",
  domain = cds$`Total_ESTAB`)

output$map.rcp <- renderLeaflet({
   if(input$count_type == 'Total_RCPTOT'){
  leaflet(data = cds) %>%
  addProviderTiles("Stamen.Toner") %>%
  addPolygons(color = ~pal(`Total_RCPTOT`), popup = ~paste0("<b>", `Industry.Title`, ":</b> ", round(`Total_RCPTOT`, 2), " $")) %>%
  addLegend(position = "bottomright", pal = pal, values = cds$`Total_RCPTOT`, title = "Value of Total Receipts filed<br> in given year (USD)")
   } else {
    leaflet(data = cds) %>%
  addProviderTiles("Stamen.Toner") %>%
  addPolygons(color = ~pal2(`Total_ESTAB`), popup = ~paste0("<b>", `Industry.Title`, ":</b> ", round(`Total_ESTAB`, 2), " estab")) %>%
  addLegend(position = "bottomright", pal = pal2, values = cds$`Total_ESTAB`, title = "Value of Total Establishments<br> in given year (USD)") 
   }
})

# pal <- colorNumeric(
#   palette = "Blues",
#   domain = cds$`Total_RCPTOT`)
# 
# output$map.rcp <- renderLeaflet({
#   leaflet(data = cds) %>%
#   addProviderTiles("Stamen.Toner") %>%
#   addPolygons(color = ~pal(`Total_RCPTOT`), popup = ~paste0("<b>", `Industry.Title`, ":</b> ", round(`Total_RCPTOT`, 2), " years")) %>%
#   addLegend(position = "bottomright", pal = pal, values = cds$`Total_RCPTOT`, title = "Value of Total Receipts filed<br> in given year (USD)")
# })

```

```{r context="render"}

leafletOutput('map.rcp')
```


### Segment Plot by Year

```{r context="server"}
unique_ind <- reactive({
    nes_st0418_p[ str_detect(nes_st0418_p$NAICS, input$naics_var) & nes_st0418_p$ST_n %in% input$ch_state & nes_st0418_p$Year %in% input$yr,]
})


output$seg.plot <- renderPlotly({
  
  if(input$count_type == 'Total_RCPTOT'){
    ggplotly(ggplot(data = unique_ind(), mapping = aes(x = reorder(Industry.Title, -Total_RCPTOT), xend = reorder(Industry.Title, -Total_RCPTOT), y = Total_RCPTOT, yend = 0, color = Industry.Title)) + 
    geom_point() +
      geom_segment()+
    coord_flip() + 
    labs(x='', y='Total Value from Receipts', title='Industry Total Value in Given Year') + theme(legend.position = "none"))
  }
  else {
    ggplotly(ggplot(data = unique_ind(), mapping = aes(x = reorder(Industry.Title, -Total_ESTAB), xend = reorder(Industry.Title, -Total_ESTAB), y = Total_ESTAB, yend = 0, color = Industry.Title)) + 
    geom_point() +
      geom_segment()+
    coord_flip() + 
    labs(x='', y='Total Number of Establishments', title='Industry Size in Given Year') + theme(legend.position = "none"))
  }
    
})


```

```{r context="render"}
#plotlyOutput("scatter.ordered")
plotlyOutput("seg.plot")

```


Column {data-width=350}
-----------------------------------------------------------------------

### Industry Details {data-width=400}

```{r}
renderTable({
  industry_subset()
})

```