PHYS 257 Lecture 6: PHYS 257 - Lecture 6 - Oct 17, 2016
Document Summary
Chi-squre helps determine how well our line of best fit represents our data. The number of degrees of freedom , k is the number of data points minus the number of parameters in your model. Note that mean of chi square distribution = k, and this has a standard deviation of sqrt(2k) For a reasonable fit the best fit" or the minimized value of 2, or 2 min is expected to be within +/- 2sigma of the mean of the chi square distribution. Recall that 95% of data on gaussian curve is within 2sigma the mean of 2 distribution is the number of fit degrees of freedom, k. This test provides a robust negative result, but not a robust positive result. Thus, the 2 statistic provides us no reason to reject the model if: This is quite large for a random number in a gaussian curve. It should fall within 2sigma of the mean.