Timeline for Why do we minimize the negative likelihood if it is equivalent to maximization of the likelihood?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Aug 28, 2020 at 23:16 | comment | added | user1389840 | Single-precision is still quite common. Double precision is not going to help much with underflow problems with large models. It will take you to 2^-1022. Log is still required either way. | |
| Aug 28, 2020 at 19:50 | comment | added | whuber♦ | Welcome to CV. Did you notice the first line of the question states "I am not asking about 'log' "? That is discussed elsewhere. BTW, it's rare for any software to perform MLE using single-precision floats anymore. | |
| Aug 28, 2020 at 18:41 | review | First posts | |||
| Aug 28, 2020 at 23:04 | |||||
| Aug 28, 2020 at 18:39 | history | answered | user1389840 | CC BY-SA 4.0 |