sugrrants: Visual methods for big temporal data

Background

This project will build a new package to support temporal data exploration and visualisation, and handle visualisation for time series models.

Related work

Packages such as forecast, hts exist for modeling time series data. They are built on specialist data structures such as ts, but temporal data typically arrives with more complications, such as irregular time, additional variables, missing values, outliers. These all need to be detected and handled before the models can be built.

Packages such as zoo provides methods for imputing missings, for regularising time series and making rolling window calculations, and provides specialist visualisation.

Packages such as tidyquant provides functions for extracting temporal data related to financial problems, and processing and plotting.

Details of your coding project

This package will provide support for plotting very long time series, slicing temporal components, new geoms such as a calendar format, and new facet systems to handle different temporal slicing.

Expected impact

Mentors, please explain how this project will produce a useful package for the R community.

Mentors

• Dr Dianne Cook, visualisation
• Dr Rob Hyndman, forecasting

Tests

• Easy: .
• Medium: .
• Hard: .

Solutions of tests

Students, please post a link to your test results here.