tprak

Theoretical Physics Practical Training
git clone git://popovic.xyz/tprak.git
Log | Files | Refs

solutions.md (1378B)


      1 # Findings
      2 
      3 ## Code
      4 The code is structured as follows:
      5 
      6     *   Construct statistical covariance matrix
      7     *   Construct Jacobi matrix of model function in terms of the parameters
      8     *   Guess parameters p0
      9 
     10 Then we iterate the following:
     11 
     12     *   Construct System covariance matrix with the guess p0
     13     *   Fill Jacobi matrix with p0 = Design Matrix
     14     *   Claculate dp
     15     *   p0 = p0 + alpha*dp; alpha \in [0, 1]
     16     *   (calculate the chi^2 function)
     17 
     18 Additionally we can calculate the chi^2 function at each iteration.
     19 
     20 The guess used for all fits was determined by standard least-square fit
     21 provided by 'scipy'. p0 = [0.9, 0.2, 0.81, 0.04, 0.02, -1, 0.84, 1.55]   # in GeV
     22 First let us look at results of single experiments fitted separately:
     23 
     24     SND -------> Pictures
     25     CMD2
     26     KLOE
     27     BABAR
     28 
     29     Table with fits and chisq, pvalue
     30 
     31 Then we can fit multiple experiments togehter if we align the given data and
     32 construct block diagonal system covariance matricies. 4 experiments fited with
     33 2 together gives 6 combinations.
     34 
     35     SND-CMD2
     36     SND-KLOE
     37     SND-BABAR
     38     CMD2-KLOE
     39     CMD2-BABAR
     40     KLOE-BABAR
     41     6 Picturs
     42 
     43     Table with fits and chisq, pvalue
     44 
     45 
     46 Here we fitted all experiments together.
     47 
     48     ALL EXPERIMENTS
     49     1 Picture
     50 
     51     Table with fits and chisq, pvalue
     52 
     53 
     54 Discussion schau ma mal was rauskommt
     55 
     56 Reference code   git://popovic.xyz/tprak.git