Within this repository is a donated project from Fall 2018 that was created in STAT 385 by Ziwei Liu, Zepeng Xiao, and Yuquan Zheng. The goal of the project was to provide guidance as to the pricing of different kinds of sneakers. Students completed this work by using:
- Web Scraping Techniques requiring both
RSelenium
andrvest
; - Manipulating data using both
tidyr
and a host of base R functions; - Modeling Sneaker Prices using
forecast
; and - Communicating the result by creating a
shiny
Application that "wraps" their work.
The package can be downloaded and used directly from GitHub with:
# Install the package from GitHub
if(!requireNamespace("devtools")) { install.packages("devtools") }
devtools::install_github("stat385uiuc/demo-sneaker-project")
# Launch the Shiny Application
sneakers::run_sneakers()
After the application is launched and a user enters in one of the pre-specified sneakers, they will see:
Within this portion of the application, the sneaker picture is displayed alongside a graph that shows the historical vs. predicted price information. Moreover, guidance is provided to both buyers and sellers as to when they should respectively buy or sell their sneakers.
If users want to compare the quantity of sneakers available across time, the comparison tab provides ample ability.
In this view, users are able to add at most two sneakers to view the quantity difference. Upon typing in the search box, they are prompted with suggested results.
Lastly, if one of the pre-cached values is not available, users can search and retrieve data using the "live search" component.
The live search portion uses RSelenium
to open an automated version of
Chrome and acquire data from a website. This can be seen by a notification
bar indicating that Chrome is under "automated test software" control.
This approach was used to retrieve and download the "pre-cached" data the application uses.