Fitting the data

The following is an example of using spf to fit data. The parameter SPF_SYSERR001 specifies the systematic error for each pulse height channel as a percentage of the count rate. The units of the model parameters are given in the spf documentation.

qcl>spf

SPF_NFILES  no. of phad files = 0 =1
SPF_PFILES001 names of pha files =mkn335_91a.phad
SPF_DFILES001 names of dmx files =$XADMX/low.idmx
SPF_SYSERR001  systematic errors for files = 0.0 =0.01
  ID masks in the LAC PHAD file are:-
 0101010101010101
 1010101010101010
SPF_LACMASK001 type 1 to fit this LAC id channel, 0 to omit = 0 =1
SPF_LACMASK002 type 1 to fit this LAC id channel, 0 to omit = 0 = 
SPF_TIME  timeslice to be fit = 1 =
SPF_MODEL model expression ='PLAW(1,2)*ABSO(3)
SPF_P001 Power law normalisation = 0.0 =1.0E-3
SPF_PE001 Power law normalisation error (0 for fixed) = 0.0 =1.0E-4
SPF_P002 Power law index (photon) = 0.0 =1.7
SPF_PE002 Power law index (photon) error (0 for fixed) = 0.0 =0.1
SPF_P003 Hydrogen column 10**21 cm-2 = 0.0 =1.0
SPF_PE003 Hydrogen column 10**21 cm-2 error (0 for fixed) = 0.0 =0.1
SPF_CONFIDENCE confidence level = 0.0 =0.9
SPF_NOINTEREST no. of uninteresting parameters = 0 =1
SPF_REDSHIFT Z for object = 0.0 =0.025
SPF_NSTEPS maximum number of iterations = 0 =50

  ID masks in the LAC PHAD file are:-
 0101010101010101
 1010101010101010

    Iteration                  Parameters             Statistic
    _________                  __________             _________

         0       0.1200E-02 1.900     10.00                  0.257E+04
  1.2000001E-03   1.900000       10.00000    
         1       0.4659E-02 1.920    0.0000E+00              0.750E+02
  4.6588276E-03   1.920209      0.0000000E+00
         2       0.4280E-02 1.924    0.0000E+00              0.397E+02
  4.2804382E-03   1.923948      0.0000000E+00
         3       0.4539E-02 1.965    0.0000E+00              0.383E+02
  4.5393999E-03   1.965073      0.0000000E+00
         4       0.4545E-02 1.965    0.0000E+00              0.383E+02
  4.5452034E-03   1.965073      0.0000000E+00
         5       0.4545E-02 1.965    0.0000E+00              0.383E+02
  4.5452029E-03   1.965073      0.0000000E+00

    Minimum value of statistic found


    Parameter    Lower bound    At minimum      Upper bound
    _________    ___________    __________      ___________

         1       0.40692E-02    0.45452E-02    0.50685E-02
         2        1.8863         1.9651         2.0447    
         3       0.00000E+00    0.00000E+00     3.1620    

    Minimum value of statistic 0.3833E+02

SPF_NAME  root name for outputs =mkn335_91a

    DX_STAT called          249 times
    Minimum reduced Chi-squared found   1.916353    
    Number of degrees of freedom           20
qcl>

Note that setting SPF_PE00n=0.0 fixes parameter "n" at the specified value. For MPC1 data, it is usually better to fit only the top-layer as the mid-layer has a lower signal to noise ratio (i.e. set SPF_LACMASK001=1 and SPF_LACMASK002=0). The program spf returns a number of parameters including the best-fit model fluxes in several energy bands (0.1-2.0 keV; 0.5-3.5 keV; 2.0-6.0 keV; 2.0-10.0 keV; 10.0-20.0 keV; 20.0-30.0 keV), the best-fit parameters, the lower and upper limits of the parameters, and the minimum reduced (see the spf documentation for further details).

To continue with the same data but a different model, then type the following:

qcl>PMODE=2
qcl>spf