Physics307L:Schedule/Week 10 agenda: Difference between revisions
From OpenWetWare
Jump to navigationJump to search
No edit summary |
No edit summary |
||
| Line 3: | Line 3: | ||
**Why fitting a line? | **Why fitting a line? | ||
***Linear relations & This versue 1/sqrt(that) | ***Linear relations & This versue 1/sqrt(that) | ||
***<math>y= | ***<math>y=Ae^{Bx}</math> | ||
**Theoretical | **Theoretical | ||
***Assume gaussian distribution for each <math>y_i</math> with same sigma for all measurements (not necessary, but simplifies it). | ***Assume gaussian distribution for each <math>y_i</math> with same sigma for all measurements (not necessary, but simplifies it). | ||
Revision as of 02:18, 22 October 2007
- Fitting a line
- AKA linear regression; least-squares fit for a line
- Why fitting a line?
- Linear relations & This versue 1/sqrt(that)
- [math]\displaystyle{ y=Ae^{Bx} }[/math]
- Theoretical
- Assume gaussian distribution for each [math]\displaystyle{ y_i }[/math] with same sigma for all measurements (not necessary, but simplifies it).
- Maximum likelihood...Chi-squared
- Derivative (minimize Chi-squared) / solve
- Just show the answer, ... skip derivation for time for practical examples.
- Practical step by step w/ Excel..."R"...."Octave?"
