data
causalis.shared.srm.NumberNumber
Value: None
None
Python Docs
causalis
Reference
Package entry
causalisRoot package overview and namespace mapNamespaces
Submodule
causalis.shared.srmsrm
Submodule causalis.shared.srm with no child pages and 12 documented members.
Classes
Jump directly into the documented classes for this page.
1 items
Functions
Jump directly into the documented functions for this page.
1 items
Data
Jump directly into the documented data for this page.
2 items
class
causalis.shared.srm.SRMResultSRMResult
Result of a Sample Ratio Mismatch (SRM) check.
Attributes
- chi2float
The calculated chi-square statistic.
- p_valuefloat
The p-value of the test, rounded to 5 decimals.
- expecteddict[Hashable, float]
Expected counts for each variant.
- observeddict[Hashable, int]
Observed counts for each variant.
- alphafloat
Significance level used for the check.
- is_srmbool
True if an SRM was detected (chi-square p-value < alpha), False otherwise.
- warningstr or None
Warning message if the test assumptions might be violated (e.g., small expected counts).
attribute
causalis.shared.srm.SRMResult.chi2chi2
Value: None
None
attribute
causalis.shared.srm.SRMResult.p_valuep_value
Value: None
None
attribute
causalis.shared.srm.SRMResult.expectedexpected
Value: None
None
attribute
causalis.shared.srm.SRMResult.observedobserved
Value: None
None
attribute
causalis.shared.srm.SRMResult.alphaalpha
Value: None
None
attribute
causalis.shared.srm.SRMResult.is_srmis_srm
Value: None
None
attribute
causalis.shared.srm.SRMResult.warningwarning
Value: None
None
method
causalis.shared.srm.SRMResult.__repr____repr__
function
causalis.shared.srm.check_srmcheck_srm
Check Sample Ratio Mismatch (SRM) for an RCT via a chi-square goodness-of-fit test.
Parameters
- assignmentsIterable[Hashable] or pandas.Series or CausalData or Mapping[Hashable, Number]
Observed variant assignments. If iterable or Series, elements are labels per unit (user_id, session_id, etc.). If CausalData is provided, the treatment column is used. If a mapping is provided, it is treated as
{variant: observed_count}with non-negative integer counts.- target_allocationdict[Hashable, Number]
Mapping
{variant: p}describing intended allocation as probabilities.- alphafloat, default 1e-3
Significance level. Use strict values like 1e-3 or 1e-4 in production.
- min_expectedfloat, default 5.0
If any expected count < min_expected, a warning is attached.
- strict_variantsbool, default True
- True: fail if observed variants differ from target keys. - False: drop unknown variants and test only on declared ones.
Returns
SRMResult
The result of the SRM check.
Raises
ValueError
If inputs are invalid or empty. ImportError If scipy is required but not installed.
Notes
Target allocation probabilities must sum to 1 within numerical tolerance.
is_srmis computed using the unrounded p-value; the returnedp_valueis rounded to 5 decimals.Missing assignments are dropped and reported via
warning.Requires SciPy for p-value computation.
Examples
Canonical target
causalis.shared.srm.check_srm
Sections
ParametersReturnsRaisesNotesExamples
data
causalis.shared.srm.__all____all__
Value: ['SRMResult', 'check_srm']
[‘SRMResult’, ‘check_srm’]