Difference between revisions of "Long Data Format"
From Open Risk Manual
Wiki admin (talk | contribs) |
Wiki admin (talk | contribs) |
||
| Line 13: | Line 13: | ||
<references /> | <references /> | ||
| − | == | + | == Implementation == |
Working with long data formats in possible in many open source python and R packages. For transition phenomena: | Working with long data formats in possible in many open source python and R packages. For transition phenomena: | ||
| − | * [https://github.com/open-risk/transitionMatrix | + | * [https://github.com/open-risk/transitionMatrix transitionMatrix Library] |
Latest revision as of 11:30, 17 May 2024
Definition
Long Data Format is a tabular representation of time series Risk Data that records the states (measurements) of multiple entities. Its defining characteristic is that each table row contains data pertaining to one entity at one point in time.
An alternative representation of the same data is the Wide Data Format.[1]
Relational Database Context
In the context of relational models of databases the combination of the Entity ID and the Event Time is the minimal set of attributes that forms a Primary Key for the relation
See Also
References
- ↑ Tidy Data, H.Wickham, Journal of Statistical Software
Implementation
Working with long data formats in possible in many open source python and R packages. For transition phenomena:
