Ignore:
File:
1 edited

Legend:

Unmodified
Added
Removed
  • src/sas/sasgui/perspectives/fitting/media/residuals_help.rst

    r84ac3f1 r99ded31  
    2727 
    2828$\chi^2$ is a statistical parameter that quantifies the differences between 
    29 an observed data set and an expected dataset (or 'theory') calculated as 
     29an observed data set and an expected dataset (or 'theory'). 
     30 
     31When showing the a model with the data, *SasView* displays this parameter 
     32normalized to the number of data points, $N_\mathrm{pts}$ such that 
    3033 
    3134.. math:: 
    3235 
    33   \chi^2 
    34   =  \sum[(Y_i - \mathrm{theory}_i)^2 / \mathrm{error}_i^2] 
     36  \chi^2_N 
     37  =  \sum[(Y_i - \mathrm{theory}_i)^2 / \mathrm{error}_i^2] / N_\mathrm{pts} 
    3538 
    36 Fitting typically minimizes the value of $\chi^2$.  For assessing the quality of 
    37 the model and its "fit" however, *SasView* displays the traditional reduced 
    38 $\chi^2_R$ which normalizes this parameter by dividing it by the number of 
    39 degrees of freedom (or DOF). The DOF is the number of data points being 
    40 considered, $N_\mathrm{pts}$, reduced by the number of free (i.e. fitted) 
    41 parameters, $N_\mathrm{par}$. Note that model parameters that are kept fixed do 
    42 *not* contribute to the DOF (they are not "free"). This reduced value is then 
    43 given as 
     39When performing a fit, *SasView* instead displays the reduced $\chi^2_R$, 
     40which takes into account the number of fitting parameters $N_\mathrm{par}$ 
     41(to calculate the number of 'degrees of freedom'). This is computed as 
    4442 
    4543.. math:: 
     
    4947  / [N_\mathrm{pts} - N_\mathrm{par}] 
    5048 
    51 Note that this means the displayed value will vary depending on the number of 
    52 parameters used in the fit. In particular, when doing a calculation without a 
    53 fit (e.g. manually changing a parameter) the DOF will now equal $N_\mathrm{pts}$ 
    54 and the $\chi^2_R$ will be the smallest possible for that combination of model, 
    55 data set, and set of parameter values. 
    56  
    57 When $N_\mathrm{pts} \gg N_\mathrm{par}$ as it should for proper fitting, the 
    58 value of the reduced $\chi^2_R$ will not change very much. 
     49The normalized $\chi^2_N$ and the reduced $\chi^2_R$ are very close to each 
     50other when $N_\mathrm{pts} \gg N_\mathrm{par}$. 
    5951 
    6052For a good fit, $\chi^2_R$ tends to 1. 
     
    9890| 2015-06-08 Steve King 
    9991| 2017-09-28 Paul Kienzle 
    100 | 2018-03-04 Paul Butler 
Note: See TracChangeset for help on using the changeset viewer.