- #1
Malamala
- 299
- 27
Hello! I really don't know much about statistics, so I am sorry if this questions is stupid or obvious. I have this data: ##x = [0,1,2,3]##, y = ##[25.885,26.139,27.404,30.230]##, ##y_{err}=[1.851,0.979,2.049,6.729]##. I need to fit to this data the following function: $$y = a (x+0.5)/4.186 + b$$ So basically a straight line (the other constants can be easily absorbed in the obtained values for a and b). I did this fit using Python and I get this values: ##a = 3.78 \pm 1.70## and ##b = 24.99 \pm 0.66## and the off diagonal value of the covariance matrix is ##-1.03##. First of all, the y values are all equal to each other, within the given errors, so I expected the value for a to be also consistent with 0 i.e. something like ##3 \pm 4##, but my value seems to be 2 sigma away from zero. Does that make sense (again please forgive my lack of knowledge in statistics)? Also, why is the covariance -1? I remember that if the variables are so well correlated is not a good sign for a fit. Does my fit makes sense? Is there anything I can do to improve it? Thank you!