The RegressionWLS companion object provides methods for setting weights and testing.
Attributes
- Companion
- class
- Graph
-
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
RegressionWLS.type
Members list
Value members
Concrete methods
Create a RegressionWLS object from a combined data-response matrix.
Create a RegressionWLS object from a combined data-response matrix.
Value parameters
- col
-
the designated response column (defaults to the last column)
- fname
-
the feature/variable names (defaults to null)
- hparam
-
the hyper-parameters (defaults to Regression.hp)
- ww
-
the weight vector (defaults to null)
- xy
-
the combined data-response matrix (predictors and response)
Attributes
Return the Root Absolute Deviations (RAD's) for each instance.
Return the Root Absolute Deviations (RAD's) for each instance.
Value parameters
- x
-
the input/data m-by-n matrix
- y
-
the response/output m-vector
Attributes
Reweight the data matrix x by multiplying by the root weight rtW.
Reweight the data matrix x by multiplying by the root weight rtW.
Value parameters
- rW
-
the root weight vector (rtW: either rootW or rW)
- x
-
the input/data m-by-n matrix
Attributes
Reweight the response vector matrix y by multiplying by the root weight rtW.
Reweight the response vector matrix y by multiplying by the root weight rtW.
Value parameters
- rW
-
the root weight vector (rtW: either rootW or rW)
- y
-
the response vector
Attributes
Estimate weights for the variables according to the reciprocal predicted rad's. Save the weight vector w and root weight vector rootW for the current model in companion object variables.
Estimate weights for the variables according to the reciprocal predicted rad's. Save the weight vector w and root weight vector rootW for the current model in companion object variables.
Value parameters
- w0
-
the initial weight vector (if null, compute it)
- x
-
the input/data m-by-n matrix
- y
-
the response/output m-vector
Attributes
Estimate weights for the variables according to the reciprocal actual rad's. Save the weight vector w and root weight vector rootW for the current model in companion object variables.
Estimate weights for the variables according to the reciprocal actual rad's. Save the weight vector w and root weight vector rootW for the current model in companion object variables.
Value parameters
- w0
-
the initial weight vector (if null, compute it)
- x
-
the input/data m-by-n matrix
- y
-
the response/output m-vector
Attributes
- See also
-
setWeights that used predicted rad
Test the given regression problem using WLS.
Test the given regression problem using WLS.
Value parameters
- w
-
the root weights
- x
-
the data matrix
- y
-
the response vector
- z
-
a vector to predict