Ontario_Canadian_EmpNAICS.Rmd

---
title: "External Canadian Emp by NAICS and Population by Province"
author: "Mausam Duggal"
date: "August 6, 2016"
output: html_document
---

```{r init, echo=FALSE, message=FALSE}
library(dplyr); library(knitr); library(gaussfacts); library(rmsfact); library(reshape2); library(rgdal); library(sp); library(stringr)

wd <- setwd("c:/personal/r")
```

#### Batch in the Ontario Business Points data that Rick cleaned 


```{r NHS}
emp <- read.csv("c:/personal/r/Ontario business points-1507.csv", stringsAsFactors = FALSE)

```

```{r}
#' Group and summarise by DAUID and then convert to wide format for getting total employment 
#' at the DA level which will be joined to the DA shapefile

da_emp <- emp %>% group_by(DAUID) %>% summarise(emp = sum(employees)) %>% .[-c(1:2),]
da_emp[is.na(da_emp)] <- 0
write.csv(da_emp, "OntarioTotEmp_DA.csv", row.names = FALSE)

```


```{r}

# Now create a long form table with the number of rows in the dataframe matching those
# of the number of employees.
da_emp1 <- emp %>% group_by(DAUID, naics2) %>% summarise(emp = sum(employees)) %>% .[-c(1:2),]
da_emp1[is.na(da_emp1)] <- 0

# create an empty dataframe
da_emp1_expanded <- as.data.frame(matrix(NA_integer_, nrow = sum(da_emp1$emp), ncol = 2))

da_emp1_expanded <- da_emp1[rep(seq(nrow(da_emp1)), da_emp1$emp), 1:2] %>% transform(., ID = seq.int(nrow(da_emp1_expanded)))
colnames(da_emp1_expanded)<- c("DAUID", "naics2", "ID")

```


```{r}
write.csv(da_emp1_expanded, "OntarioEmp_DA.csv", row.names = FALSE)

```