I had some time this past weekend so I created a Dash App to deploy on Heroku as part of this website portfolio. I’ve created visualizations in plotly using both R and Python but I wanted to show off my Dash skills with a relatively simple application that I could walk through in a short amount of time. Here’s a screen-cap preview:
Screen capture of Gas App hosted on Heroku
The Tableau dashboard software is a truly powerful visualization tool. Using their Public Verison, I was able to create a very powerful dashboard visual in about an hour.
The paid version at the time of this post is $70/month but if you don’t mind saving to their public site, the free option is just fine. Don’t worry if you’re just starting out: Tableau offers users the option of hiding visuals and data from view if you’re not ready go public.
Traditional forecasting methods require stationary data to make a valid forecast. In this post I’ll run through a relatively simple time-series forecast that employs forcast package’s auto.arima function and regression tress to via the rpart library.
library(RCurl)
library(forecast)
library(lubridate)
library(tidyverse)
library(ggforce)
library(kableExtra)
library(rpart)
library(rpart.plot)
library(data.table)
Here’s a preview of the data I’m working with. It’s the same EMS data (with different aggregations) from my EMS Flexdashboard post.
https://jbrnbrg.
I’ve recently deployed a flexdashboard on shinyapps.io that allows a user to compare the EMS call counts and dispatch/response times between boroughs of NYC. You can check out the application here.
EMS Flexdashboard Screencap
The data used by the dashboard is read directly from the NYC Open data API and it covers each call for all 5 boroughs from 2013-2017 inclusive. The data contains 7.2M individual observations (emergency calls into the system) with a variety of features including call type, severity level, borough, and zip code.