| | |
- MultivarRegr(data=None, data_filter_expr=None, variables_count=None, weights=False, formula=None, algorithm=None, coeff_stats=False, conf_int_level=0.9, model_stats=False, residuals=False, **generic_arguments)
- DESCRIPTION:
The MultivarRegr() function is a multivariate linear regression function.
Using a formula that defines the relationship between the explanatory
variable and multiple response variables, it fits data to a multidimensional surface.
PARAMETERS:
data:
Required Argument.
Specifies series with payload characteristics as follows:
* Payload content value is MULTIVAR_REAL.
* Payload field value has "variables_count" required fields (response
variable followed by explanatory variables).
Types: TDSeries
data_filter_expr:
Optional Argument.
Specifies filter expression for "data".
Types: ColumnExpression
variables_count:
Required Argument.
Specifies how many parameters are present in the payload. For
linear regression with no weighting, there are 2 variables (the response
variable and the explanatory variable). For linear regression with weighting,
there are 3 variables (the response variable, the explanatory variable, and
the weights).
Types: int
weights:
Optional Argument.
Specifies whether a third series is present in MULTIVAR series-specifications.
The third series is interpreted as a series of weights that can be used to
perform a weighted least-squares regression solution.
When set to False, no third series is present, otherwise it is present.
Default Value: False
Types: bool
formula:
Required Argument.
Specifies the relationship between the explanatory variable and the response variables.
Types: str
algorithm:
Required Argument.
Specifies algorithm used for the regression.
Permitted Values:
1. QR: means that QR decomposition is used for the regression.
2. PSI: means that pseudo-inverse based on singular value
decomposition (SVD) is used to solve the regression.
Types: str
coeff_stats:
Optional Argument.
Specifies whether to include coefficient statistics columns in the output or not.
When set to False, coefficient statistics columns are not returned in the output,
otherwise columns are returned in the output.
Default Value: False
Types: bool
conf_int_level:
Optional Argument.
Specifies the confidence interval level value used for coefficient statistics
calculation.
The value should be greater than 0 and less than 1.
Note:
Applicable only when "coeff_stats" is set to 1.
Default Value: 0.9
Types: float
model_stats:
Optional Argument.
Specifies whether to generate the optional model statistics.
The generated result set can be retrieved using the attribute
fitmetadata of the function output. When set to False, function
does not generate the statistics, otherwise generates it.
Default Value: False
Types: bool
residuals:
Optional Argument.
Specifies whether to generate the tertiary (residuals) layer.
The generated result set can be retrieved using the attribute
fitresiduals of the function output. when set to False, function
does not generate the residual data, otherwise generates it.
Default Value: False
Types: bool
**generic_arguments:
Specifies the generic keyword arguments of UAF functions.
Below are the generic keyword arguments:
persist:
Optional Argument.
Specifies whether to persist the results of the
function in a table or not. When set to True,
results are persisted in a table; otherwise,
results are garbage collected at the end of the
session.
Note that, when UAF function is executed, an
analytic result table (ART) is created.
Default Value: False
Types: bool
volatile:
Optional Argument.
Specifies whether to put the results of the
function in a volatile ART or not. When set to
True, results are stored in a volatile ART,
otherwise not.
Default Value: False
Types: bool
output_table_name:
Optional Argument.
Specifies the name of the table to store results.
If not specified, a unique table name is internally
generated.
Types: str
output_db_name:
Optional Argument.
Specifies the name of the database to create output
table into. If not specified, table is created into
database specified by the user at the time of context
creation or configuration parameter. Argument is ignored,
if "output_table_name" is not specified.
Types: str
RETURNS:
Instance of MultivarRegr.
Output teradataml DataFrames can be accessed using attribute
references, such as MultivarRegr_obj.<attribute_name>.
Output teradataml DataFrame attribute names are:
1. result
2. fitmetadata - Available when "model_stats" is set to True, otherwise not.
3. fitresiduals - Available when "residuals" is set to True, otherwise not.
RAISES:
TeradataMlException, TypeError, ValueError
EXAMPLES:
# Notes:
# 1. Get the connection to Vantage to execute the function.
# 2. One must import the required functions mentioned in
# the example from teradataml.
# 3. Function will raise error if not supported on the Vantage
# user is connected to.
# Check the list of available UAF analytic functions.
display_analytic_functions(type="UAF")
# Load the example data.
load_example_data("uaf", ["house_values"])
# Create teradataml DataFrame object.
df = DataFrame.from_table("house_values")
# Create teradataml TDSeries object.
data_series_df = TDSeries(data=df,
id="cityid",
row_index="TD_TIMECODE",
payload_field=["house_val","salary","mortgage"],
payload_content="MULTIVAR_REAL")
# Example 1 : Execute multivariate regression function to identify the degree of
# linearity the explanatory variable and multiple response variables.
# Generate the model statistics and residual data as well.
uaf_out = MultivarRegr(data=data_series_df,
variables_count=3,
weights=False,
formula="Y = B0 + B1*X1 + B2*X2",
algorithm='QR',
coeff_stats=True,
model_stats=True,
residuals=True)
# Print the result DataFrames.
print(uaf_out.result)
print(uaf_out.fitmetadata)
print(uaf_out.fitresiduals)
|