cochrane_orcutt¶

In [ ]:

cochrane_orcutt(model, 
                vdf: (vDataFrame, str), 
                ts: str, 
                prais_winsten: bool = False, 
                drop_tmp_model: bool = True,)

Performs a Cochrane-Orcutt estimation.

Parameters¶

Name	Type	Optional	Description
model	vModel	❌	Linear regression object.
vdf	vDataFrame / str	❌	Input relation.
ts	str	❌	vcolumn of numeric or date-like type (date, datetime, timestamp, etc.) used as the timeline and to order the data.
prais_winsten	bool	✓	If true, retains the first observation of the time series, increasing precision and efficiency. This configuration is called the Prais–Winsten estimation.
drop_tmp_model	bool	✓	If true, it drops the temporary model.

Returns¶

model

A linear model with the following information stored as attributes:

coef_: Model's coefficients.
pho_: Cochrane-Orcutt pho.
anova_table_: ANOVA table.
r2_: R2.

Example¶

In [15]:

from verticapy.datasets import load_amazon
amazon = load_amazon().search("state = 'ACRE'")
amazon["number_bias"] = "POWER(number, 2) - 500 + RANDOM() * 1000"

from verticapy.learn.linear_model import LinearRegression
model = LinearRegression("model_lr")
model.drop()
model.fit(amazon, ["number_bias"], "number")
model.coef_

Out[15]:

	Abc predictor Varchar(65000)	123 coefficient Float	123 std_err Float	123 t_value Float	123 p_value Float
1	Intercept	119.646769123412	18.0214014796031	6.63914897289369	2.13193113743317e-10
2	number_bias	0.000287117395412203	7.15827828366076e-06	40.1098398294416	1.28381262117036e-107

Rows: 1-2 | Columns: 5

In [17]:

from verticapy.stats import cochrane_orcutt
co = cochrane_orcutt(model, amazon, "date")
co.coef_

Out[17]:

	Abc predictor Varchar(65000)	123 coefficient Float	123 std_err Float	123 t_value Float	123 p_value Float
1	Intercept	82.9974261969997	16.0873800250863	5.15916364675761	5.25690500077592e-07
2	number_bias	0.000266082574639292	6.91739924957652e-06	38.4656957100721	1.22772718982409e-103

Rows: 1-2 | Columns: 5

In [18]:

co.pho_

Out[18]:

0.381523480423635