NRAO Home > CASA > CASA Toolkit Reference Manual
ms.statwt - Function

1.3.1 Set WEIGHT and SIGMA from the scatter of the visibilities


Description

NOT IMPLEMENTED YET.

This function estimates the noise from the scatter of the visibilities, sets SIGMA to it, and WEIGHT to SIGMA**-2.

Ideally the visibilities used to estimate the scatter, as selected by fitspw and fitcorr, should be pure noise. If you know for certain that they are, then setting dorms to True will give the best result. Otherwise, use False (standard sample standard deviation). More robust scatter estimates like the interquartile range or median absolute deviation from the median are not offered because they require sorting by value, which is not possible for complex numbers.

To beat down the noise of the noise estimate, the sample size per estimate can be made larger than a single spw and baseline. (Using combine=’spw’ is to interpolate between spws with line-free channels is recommended when an spw has no line-free channels.) timebin smooths the noise estimate over time. windowtype sets the type of time smoothing.

WEIGHT and SIGMA will not be changed for samples that have fewer than minsamp visibilities. Selected visibilities for which no noise estimate is made will be flagged. Note that minsamp is effectively at least 2 if dorms is False, and 1 if it is True.

Arguments





Inputs

dorms

How the scatter should be estimated (True -> rms, False -> stddev)

allowed:

bool

Default:

false

byantenna

How the scatters are solved for (by antenna or by baseline)

allowed:

bool

Default:

true

sepacs

If solving by antenna, treat autocorrs separately

allowed:

bool

Default:

true

fitspw

Line-free spectral windows (and :channels) to get the scatter from. (” => all)

allowed:

variant

Default:

*

fitcorr

Correlations (V, LL, XX, LR, XY, etc.) to get the scatter from. (” => all)

allowed:

variant

Default:

combine

Ignore changes in these columns (spw, scan, and/or state) when getting the scatter.

allowed:

string

Default:

timebin

Duration of the moving window over which to estimate the scatter. Defaults to 0s, with an effective minimum of 1 integration.

allowed:

variant

Default:

0s

minsamp

The minimum number of visibilities for a scatter estimate

allowed:

int

Default:

3

field

Fields to reweight, by names or 0-based ids. (” => all)

allowed:

variant

Default:

spw

Spectral windows to reweight. (” => all)

allowed:

variant

Default:

*

antenna

Select data based on antenna/baseline

allowed:

any

Default:

variant

timerange

Select data by time range

allowed:

string

Default:

scan

Scan numbers to reweight. (” => all)

allowed:

variant

Default:

intent

Scan intents to reweight. (” => all)

allowed:

string

Default:

array

Select (sub)array(s) by array ID number

allowed:

variant

Default:

correlation

Correlations (LL, XX, LR, XY, etc.) to reweight. (” => all)

allowed:

string

Default:

obs

Observation IDs to reweight. (” => all)

allowed:

string

Default:

datacolumn

Which data column to calculate the scatter from.

allowed:

string

Default:

corrected_data

Returns
bool

Example

 
 
  ms.open("multiwin.ms", nomodify=False)  
  ms.statwt(fitspw=’0:0~123;145~211,2:124~255’, field=[0], spw=’0,2’)  
 
  In this example the noise estimates are seperately made from and applied  
  to spws 0 and 2.  
 
  ms.statwt(fitspw=’0:0~123;145~211,2:124~255’, fitorder=0, field=[0],  
            combine=’spw’)  
  ms.close()  
 
  This time the estimate for each baseline is made from the line-free channels  
  of spws 0 and 2, and applied to all the spws, including 1 (which could be a  
  completely line-filled spw).  
 

__________________________________________________________________


More information about CASA may be found at the CASA web page

Copyright 2016 Associated Universities Inc., Washington, D.C.

This code is available under the terms of the GNU General Public Lincense


Home | Contact Us | Directories | Site Map | Help | Privacy Policy | Search