SyntheticDiffInDiffEstimator implements SyntheticDiffInDiffEstimator More...

Inheritance diagram for Synapse.ML.Causal.SyntheticDiffInDiffEstimator:

[legend]

Collaboration diagram for Synapse.ML.Causal.SyntheticDiffInDiffEstimator:

[legend]

Public Member Functions
	SyntheticDiffInDiffEstimator ()
	Creates a SyntheticDiffInDiffEstimator without any parameters. More...

	SyntheticDiffInDiffEstimator (string uid)
	Creates a SyntheticDiffInDiffEstimator with a UID that is used to give the SyntheticDiffInDiffEstimator a unique ID. More...

SyntheticDiffInDiffEstimator	SetEpsilon (double value)
	Sets value for epsilon More...

SyntheticDiffInDiffEstimator	SetHandleMissingOutcome (string value)
	Sets value for handleMissingOutcome More...

SyntheticDiffInDiffEstimator	SetLocalSolverThreshold (long value)
	Sets value for localSolverThreshold More...

SyntheticDiffInDiffEstimator	SetMaxIter (int value)
	Sets value for maxIter More...

SyntheticDiffInDiffEstimator	SetNumIterNoChange (int value)
	Sets value for numIterNoChange More...

SyntheticDiffInDiffEstimator	SetOutcomeCol (string value)
	Sets value for outcomeCol More...

SyntheticDiffInDiffEstimator	SetPostTreatmentCol (string value)
	Sets value for postTreatmentCol More...

SyntheticDiffInDiffEstimator	SetStepSize (double value)
	Sets value for stepSize More...

SyntheticDiffInDiffEstimator	SetTimeCol (string value)
	Sets value for timeCol More...

SyntheticDiffInDiffEstimator	SetTol (double value)
	Sets value for tol More...

SyntheticDiffInDiffEstimator	SetTreatmentCol (string value)
	Sets value for treatmentCol More...

SyntheticDiffInDiffEstimator	SetUnitCol (string value)
	Sets value for unitCol More...

SyntheticDiffInDiffEstimator	SetZeta (double value)
	Sets value for zeta More...

double	GetEpsilon ()
	Gets epsilon value More...

string	GetHandleMissingOutcome ()
	Gets handleMissingOutcome value More...

long	GetLocalSolverThreshold ()
	Gets localSolverThreshold value More...

int	GetMaxIter ()
	Gets maxIter value More...

int	GetNumIterNoChange ()
	Gets numIterNoChange value More...

string	GetOutcomeCol ()
	Gets outcomeCol value More...

string	GetPostTreatmentCol ()
	Gets postTreatmentCol value More...

double	GetStepSize ()
	Gets stepSize value More...

string	GetTimeCol ()
	Gets timeCol value More...

double	GetTol ()
	Gets tol value More...

string	GetTreatmentCol ()
	Gets treatmentCol value More...

string	GetUnitCol ()
	Gets unitCol value More...

double	GetZeta ()
	Gets zeta value More...

override DiffInDiffModel	Fit (DataFrame dataset)
	Fits a model to the input data. More...

void	Save (string path)
	Saves the object so that it can be loaded later using Load. Note that these objects can be shared with Scala by Loading or Saving in Scala. More...

JavaMLWriter	Write ()
	Returns a JavaMLWriter instance for this ML instance.

JavaMLReader< SyntheticDiffInDiffEstimator >	Read ()
	Get the corresponding JavaMLReader instance. More...

Static Public Member Functions
static SyntheticDiffInDiffEstimator	Load (string path)
	Loads the SyntheticDiffInDiffEstimator that was previously saved using Save(string). More...

Detailed Description

SyntheticDiffInDiffEstimator implements SyntheticDiffInDiffEstimator

Constructor & Destructor Documentation

◆ SyntheticDiffInDiffEstimator() [1/2]

Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SyntheticDiffInDiffEstimator ( )

inline

Creates a SyntheticDiffInDiffEstimator without any parameters.

◆ SyntheticDiffInDiffEstimator() [2/2]

Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SyntheticDiffInDiffEstimator ( string uid )

inline

Creates a SyntheticDiffInDiffEstimator with a UID that is used to give the SyntheticDiffInDiffEstimator a unique ID.

Parameters

uid	An immutable unique ID for the object and its derivatives.

Member Function Documentation

◆ Fit()

override DiffInDiffModel Synapse.ML.Causal.SyntheticDiffInDiffEstimator.Fit ( DataFrame dataset )

Fits a model to the input data.

Parameters

dataset The DataFrame to fit the model to.

Returns: DiffInDiffModel

◆ GetEpsilon()

double Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetEpsilon ( )

Gets epsilon value

Returns: epsilon: This value is added to the weights when we fit the final linear model for SyntheticControlEstimator and SyntheticDiffInDiffEstimator in order to avoid zero weights.

◆ GetHandleMissingOutcome()

string Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetHandleMissingOutcome ( )

Gets handleMissingOutcome value

Returns: handleMissingOutcome: How to handle missing outcomes. Options are skip (which will filter out units with missing outcomes), zero (fill in missing outcomes with zero), or impute (impute with nearest available outcomes, or mean if two nearest outcomes are available)

◆ GetLocalSolverThreshold()

long Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetLocalSolverThreshold ( )

Gets localSolverThreshold value

Returns: localSolverThreshold: threshold for using local solver on driver node. Local solver is faster but relies on part of data being collected on driver node.

◆ GetMaxIter()

int Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetMaxIter ( )

Gets maxIter value

Returns: maxIter: maximum number of iterations (>= 0)

◆ GetNumIterNoChange()

int Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetNumIterNoChange ( )

Gets numIterNoChange value

Returns: numIterNoChange: Early termination when number of iterations without change reached.

◆ GetOutcomeCol()

string Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetOutcomeCol ( )

Gets outcomeCol value

Returns: outcomeCol: outcome column

◆ GetPostTreatmentCol()

string Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetPostTreatmentCol ( )

Gets postTreatmentCol value

Returns: postTreatmentCol: post treatment indicator column

◆ GetStepSize()

double Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetStepSize ( )

Gets stepSize value

Returns: stepSize: Step size to be used for each iteration of optimization (> 0)

◆ GetTimeCol()

string Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetTimeCol ( )

Gets timeCol value

Returns: timeCol: Specify the column that identifies the time when outcome is measured in the panel data. For example, if the outcome is measured daily, this column could be the Date column.

◆ GetTol()

double Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetTol ( )

Gets tol value

Returns: tol: the convergence tolerance for iterative algorithms (>= 0)

◆ GetTreatmentCol()

string Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetTreatmentCol ( )

Gets treatmentCol value

Returns: treatmentCol: treatment column

◆ GetUnitCol()

string Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetUnitCol ( )

Gets unitCol value

Returns: unitCol: Specify the name of the column which contains an identifier for each observed unit in the panel data. For example, if the observed units are users, this column could be the UserId column.

◆ GetZeta()

double Synapse.ML.Causal.SyntheticDiffInDiffEstimator.GetZeta ( )

Gets zeta value

Returns: zeta: The zeta value for regularization term when fitting unit weights. If not specified, a default value will be computed based on formula (2.2) specified in https://www.nber.org/system/files/working_papers/w25532/w25532.pdf. For large scale data, one may want to tune the zeta value, minimizing the loss of the unit weights regression.

◆ Load()

static SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.Load ( string path )

static

Loads the SyntheticDiffInDiffEstimator that was previously saved using Save(string).

Parameters

path	The path the previous SyntheticDiffInDiffEstimator was saved to

Returns: New SyntheticDiffInDiffEstimator object, loaded from path.

◆ Read()

JavaMLReader<SyntheticDiffInDiffEstimator> Synapse.ML.Causal.SyntheticDiffInDiffEstimator.Read ( )

Get the corresponding JavaMLReader instance.

Returns: an JavaMLReader<SyntheticDiffInDiffEstimator> instance for this ML instance.

◆ Save()

void Synapse.ML.Causal.SyntheticDiffInDiffEstimator.Save ( string path )

Saves the object so that it can be loaded later using Load. Note that these objects can be shared with Scala by Loading or Saving in Scala.

Parameters

path	The path to save the object to

◆ SetEpsilon()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetEpsilon ( double value )

Sets value for epsilon

Parameters

value This value is added to the weights when we fit the final linear model for SyntheticControlEstimator and SyntheticDiffInDiffEstimator in order to avoid zero weights.

Returns: New SyntheticDiffInDiffEstimator object

◆ SetHandleMissingOutcome()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetHandleMissingOutcome ( string value )

Sets value for handleMissingOutcome

Parameters

value How to handle missing outcomes. Options are skip (which will filter out units with missing outcomes), zero (fill in missing outcomes with zero), or impute (impute with nearest available outcomes, or mean if two nearest outcomes are available)

Returns: New SyntheticDiffInDiffEstimator object

◆ SetLocalSolverThreshold()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetLocalSolverThreshold ( long value )

Sets value for localSolverThreshold

Parameters

value threshold for using local solver on driver node. Local solver is faster but relies on part of data being collected on driver node.

Returns: New SyntheticDiffInDiffEstimator object

◆ SetMaxIter()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetMaxIter ( int value )

Sets value for maxIter

Parameters

value maximum number of iterations (>= 0)

Returns: New SyntheticDiffInDiffEstimator object

◆ SetNumIterNoChange()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetNumIterNoChange ( int value )

Sets value for numIterNoChange

Parameters

value Early termination when number of iterations without change reached.

Returns: New SyntheticDiffInDiffEstimator object

◆ SetOutcomeCol()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetOutcomeCol ( string value )

Sets value for outcomeCol

Parameters

value outcome column

Returns: New SyntheticDiffInDiffEstimator object

◆ SetPostTreatmentCol()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetPostTreatmentCol ( string value )

Sets value for postTreatmentCol

Parameters

value post treatment indicator column

Returns: New SyntheticDiffInDiffEstimator object

◆ SetStepSize()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetStepSize ( double value )

Sets value for stepSize

Parameters

value Step size to be used for each iteration of optimization (> 0)

Returns: New SyntheticDiffInDiffEstimator object

◆ SetTimeCol()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetTimeCol ( string value )

Sets value for timeCol

Parameters

value Specify the column that identifies the time when outcome is measured in the panel data. For example, if the outcome is measured daily, this column could be the Date column.

Returns: New SyntheticDiffInDiffEstimator object

◆ SetTol()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetTol ( double value )

Sets value for tol

Parameters

value the convergence tolerance for iterative algorithms (>= 0)

Returns: New SyntheticDiffInDiffEstimator object

◆ SetTreatmentCol()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetTreatmentCol ( string value )

Sets value for treatmentCol

Parameters

value treatment column

Returns: New SyntheticDiffInDiffEstimator object

◆ SetUnitCol()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetUnitCol ( string value )

Sets value for unitCol

Parameters

value Specify the name of the column which contains an identifier for each observed unit in the panel data. For example, if the observed units are users, this column could be the UserId column.

Returns: New SyntheticDiffInDiffEstimator object

◆ SetZeta()

SyntheticDiffInDiffEstimator Synapse.ML.Causal.SyntheticDiffInDiffEstimator.SetZeta ( double value )

Sets value for zeta

Parameters

value The zeta value for regularization term when fitting unit weights. If not specified, a default value will be computed based on formula (2.2) specified in https://www.nber.org/system/files/working_papers/w25532/w25532.pdf. For large scale data, one may want to tune the zeta value, minimizing the loss of the unit weights regression.

Returns: New SyntheticDiffInDiffEstimator object

The documentation for this class was generated from the following file:

synapse/ml/causal/SyntheticDiffInDiffEstimator.cs

Public Member Functions

Static Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ SyntheticDiffInDiffEstimator() [1/2]

◆ SyntheticDiffInDiffEstimator() [2/2]

Member Function Documentation

◆ Fit()

◆ GetEpsilon()

◆ GetHandleMissingOutcome()

◆ GetLocalSolverThreshold()

◆ GetMaxIter()

◆ GetNumIterNoChange()

◆ GetOutcomeCol()

◆ GetPostTreatmentCol()

◆ GetStepSize()

◆ GetTimeCol()

◆ GetTol()

◆ GetTreatmentCol()

◆ GetUnitCol()

◆ GetZeta()

◆ Load()

◆ Read()

◆ Save()

◆ SetEpsilon()

◆ SetHandleMissingOutcome()

◆ SetLocalSolverThreshold()

◆ SetMaxIter()

◆ SetNumIterNoChange()

◆ SetOutcomeCol()

◆ SetPostTreatmentCol()

◆ SetStepSize()

◆ SetTimeCol()

◆ SetTol()

◆ SetTreatmentCol()

◆ SetUnitCol()

◆ SetZeta()