Outlier: Difference between revisions
From ACT Wiki
Jump to navigationJump to search
imported>Doug Williamson (Create page. Sources: Linked pages.) |
imported>Doug Williamson (Add link.) |
||
| (One intermediate revision by the same user not shown) | |||
| Line 30: | Line 30: | ||
*[[Exception]] | *[[Exception]] | ||
*[[Hedging]] | *[[Hedging]] | ||
*[[Machine learning]] (ML) | |||
*[[MD]] | *[[MD]] | ||
*[[Neural network]] | |||
*[[Reporting]] | *[[Reporting]] | ||
*[[Trading]] | *[[Trading]] | ||
Latest revision as of 09:25, 28 September 2022
1. Data.
A data point that is much larger, or much smaller, than the next nearest data point.
2. Controls.
An unusual item that warrants investigation.
- Machine learning (ML) algorithms can find outliers
- "More will become possible in the future, as ML capabilities are increasingly applied in treasury, because of their strengths in data analysis, finding patterns and solving difficult mathematical problems.
- 'A TMS powered by ML can suggest alternatives for trading, hedging or optimising the capital structure.
- ML algorithms can find outliers and alarm the user in areas such as anti-fraud, sanction screening, security and operations,' says Philipp Leitner, co-CTO and MD at ION Treasury."
- The Treasurer - 2022 Issue 3 - p8.
