Overview¶
The main class that is responsible for handling the data statistics, as
well as the covariance matrix used during parameter estimation, is
the pyRSD.rsdfit.data.PowerData
class.
Information about the parameters that are needed to initialize
the PowerData
class can be found by using
the PowerData.help()
function,
In [1]: from pyRSD.rsdfit.data import PowerData
# print out the help message for the parameters needed
In [2]: PowerData.help()
Initialization Parameters for PowerData
--------------------------------------------------
covariance :
The string specifying the name of the file holding the covariance matrix.
covariance_Nmocks :
The number of mocks that was used to measure the covariance matrix.
If this is non-zero, then the inverse covariance matrix will
be rescaled to account for noise due to the finite number of mocks
Default: 0.0
covariance_rescaling :
Rescale the covariance matrix read from file by this amount.
Default: 1.0
data_file :
The string specifying the name of the file holding the data measurements.
ells :
A list of integers specifying multipole numbers for each statistic
in the final analysis.
This must be supplied when the :attr:`mode` is ``poles``
Default: None
fitting_range :
The :math:`k` fitting range for each statistics.
This can either be a tuple of (kmin, kmax), which will be
used for each statistic or a list of tuples of (kmin, kmax)
grid_file :
A string specifying the name of the file holding a
:class:`pyRSD.rsd.transfers.PkmuGrid` to read.
Default: None
max_ellprime :
When convolving a multipole of order ``ell``, include contributions
up to and including this number.
Default: 4
mode :
The type of data, either ``pkmu`` or ``poles``
mu_bounds :
A list of tuples specifying the edges of the :math:`\mu` bins.
This should have (mu_min, mu_max), corresponding
to the edges of the bins for each statistic in the final analysis
This must be supplied when the :attr:`mode` is ``pkmu``
Default: None
statistics :
A list of the string names for each statistic that will be read from file
These strings should be of the form:
>> ['pole_0', 'pole_2', ...]
>> ['pkmu_0.1', 'pkmu_0.3', ...]
usedata :
A list of the statistic numbers that will be included in the final
analysis.
This allows the user to exclude certain statistics read from file. By
default (``None``), all statistics are included
Default: None
window_file :
A string specifying the name of the file holding the correlation
function multipoles of the window function.
The file should contain columns of data, with the first column
specifying the separation array :math:`s`, and the other columns
giving the even-numbered correlation function multipoles of the window
Default: None
window_kmax :
Default kmax value to use on the grid when convolving the model.
Default: 0.7
window_kmin :
Default kmin value to use on the grid when convolving the model.
Default: 0.0001
These parameters should be specified in the parameter file that is passed
to the rsdfit
executable and the names of the parameters should be
prefixed with the data.
prefix. In our example parameter file
discussed previously, we specify multipoles data to read from file as
#-------------------------------------------------------------------------------
# data params
#-------------------------------------------------------------------------------
data.covariance = '$(PYRSD_DATA)/examples/runPB_poles_gaussian_cov.dat'
data.covariance_Nmocks = 0
data.covariance_rescaling = 1.0
data.data_file = '$(PYRSD_DATA)/examples/runPB_galaxy_poles.dat'
data.ells = [0, 2, 4]
data.fitting_range = [(0.02, 0.4), (0.02, 0.4), (0.02, 0.4)]
data.grid_file = '$(PYRSD_DATA)/examples/runPB_pkmu_grid.dat'
data.mode = 'poles'
data.mu_bounds = None
data.statistics = ['pole_0', 'pole_2', 'pole_4']
data.usedata = range(0, 3)
data.window_file = None
#-------------------------------------------------------------------------------
These parameters allow the user to specify which type of data is being
used by specifying the mode
parameter, either pkmu
for \(P(k,\mu)\)
data or poles
for \(P_\ell(k)\) data. The user can also specify the
desired \(k\) ranges to use when fitting the data, via the
fitting_range
parameter.
The data itself must be read in from a plaintext file. Similarly, the covariance matrix and grid file must also be read from a plaintext file. See the next section File Formats for Data Files for more specifics on the format of these plaintext files.