One of the very first programming “tricks” I learned was that you could compute both the mean and the standard deviation with a single pass through the data values. However, the single-pass expression leads to larger roundoff errors than the two-pass expression, which limits its usefulness.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Array intrinsics performances/accuracy
|
20 | 874 | May 20, 2023 | |
Speed of array intrinsics
|
22 | 1900 | April 14, 2025 | |
Fast_math: A collection of functions for fast number crunching using Fortran | 12 | 1206 | October 3, 2023 | |
Use case of single-precision real number | 20 | 1184 | March 17, 2024 | |
Speed of computing matmul(A, transpose(A))
|
2 | 663 | June 6, 2023 |