Granger Causality Operator

operator.

/opt/hostedtoolcache/Python/3.10.12/x64/lib/python3.10/site-packages/fastcore/docscrape.py:225: UserWarning: Unknown section Notes
  else: warn(msg)
/opt/hostedtoolcache/Python/3.10.12/x64/lib/python3.10/site-packages/fastcore/docscrape.py:225: UserWarning: Unknown section See Also
  else: warn(msg)

---

source

GrangerCausality

 GrangerCausality (x_vars:Union[list,pandas.core.series.Series,numpy.ndarr
                   ay,Iterable[numbers.Number],Any,List[int],List[Union[bo
                   ol,numpy.bool_,Literal[0],Literal[1]]],pandas.core.inde
                   xes.base.Index,NoneType]=None, y_vars:Union[list,pandas
                   .core.series.Series,numpy.ndarray,Iterable[numbers.Numb
                   er],Any,List[int],List[Union[bool,numpy.bool_,Literal[0
                   ],Literal[1]]],pandas.core.indexes.base.Index,NoneType]
                   =None, shift:Optional[int]=10,
                   test:Optional[str]='ssr_chi2test',
                   lag_order:Optional[int]=1,
                   max_lag:Optional[tuple]=(1,), n_jobs:Optional[int]=-1)

Computes Granger Causality Check Granger Causality of all possible combinations of the Time series.

The rows are the response variable, columns are predictors. The values in the table are the P-Values. P-Values lesser than the significance level (0.05), implies the Null Hypothesis that the coefficients of the corresponding past values is zero, that is, the X does not cause Y can be rejected.

	Type	Default	Details
x_vars	Union	None	A subset of response variable (features) to compute granger’s causality test with. If not provided, defaults to df.index.values i.e. all rows in df.
y_vars	Union	None	A subset of response variable (features) to compute granger’s causality test with. If not provided, defaults to df.index.values i.e. all rows in df.
shift	Optional	10	number to shift df’s values by
test	Optional	ssr_chi2test	the kind of statistical test to use
lag_order	Optional	1	how long to lag
max_lag	Optional	(1,)	if None coerced to (1, )
n_jobs	Optional	-1	number of cpu threads to use during calculation
Returns	pd.DataFrame		Pandas DataFrame with shape (len(x_vars), len(y_vars)) containing the minimum p-value from Granger’s Causation Tests