Thomson-Reuters Paint

Inspiration

Financial data is often very complex and hard to get an intuitive understanding of easily, especially when it comes to time series data. So we set out to create a easy to use application that could visualize some aspects of financial data so users of all levels could gain intuition through it by the simple clicks of a mouse.

What it does

It enables the user to select several preconstructed portfolio of stocks, and then select two parameters out of a possible 6 for visualization. Then it would plot the data against an axis determined by the selected parameters. The user could then adjust the slider to see the changes in the selected parameters over time.

How we built it

We built it using R. We used the httr package to post JSON queries using the Thomson-Reuters API key provided, and used the jsonlite package to process and parse the JSON response into R data format. Then we used the r base plot function to build the plot, and implemented the entire program using the R Shiny GUI framework to enable interaction.

Challenges we ran into

The JSON results were really hard to parse, and the Shiny structure is pretty difficult to understand at first.

Accomplishments that we're proud of

Not fall asleep at the same time during the night.

What we learned

Coding is hard.

What's next for Thomson-Reuters Paint

Based on our prototype, we have shown how time series financial data could be displayed in a simple and intuitive manner. However due to time constraints and decreasing productivity as we became more and more sleep deprived, we did chose to not implement many functionalities during development. For example, we did not allow the user to manually enter in a portfolio, and we also fixed the number of time steps for the time step slider. These could both be customized if given enough time. Also, we built the plot using R's base graphic functions, but we could have definitely used ggplot2 to build the plot. If given enough time these functionalities could all be implemented.