object Variable extends Error with Serializable
The Variable
object provides utilities for variables.
- Alphabetic
- By Inheritance
- Variable
- Serializable
- Serializable
- Error
- AnyRef
- Any
- Hide All
- Show All
- Public
- All
Value Members
- val VAL0: Double
- val VAL1: Double
-
def
dummyVar(tt: Int, shf: Int = shift, tmx: Int = tmax): VectoD
Assign values for dummy variables based on a single categorical/treatment value 'tt'.
Assign values for dummy variables based on a single categorical/treatment value 'tt'.
- tt
the categorical/treatment value
- tmx
the maximum categorical/treatment after shifting
-
def
dummyVars(tt: VectoI): MatriD
Assign values for the dummy variables based on the categorical/treatment vector 'tt'.
Assign values for the dummy variables based on the categorical/treatment vector 'tt'. A single categorical variable 'tt' with values 'ttmin' to 'ttmax' will be (1) shifted to the range 0 to 'tmax' and then replace by 'tmax' dummy variables/columns as follows:
0 => 0, 0, 0 OR 1, 1, 1 1 => 1, 0, 0 OR 2, 1, 1 2 => 0, 1, 0 OR 1, 2, 1 3 => 0, 0, 1 OR 1, 1, 2
Using (0, 1) for (VAL0, VAL1) is conventional, but using (1, 2) reduces collinearity, for example in
QuadRegression
. Note: one-hot encoding using 3 dummy variable leads to singular matrices.- tt
the categorical/treatment vector
-
final
def
flaw(method: String, message: String): Unit
- Definition Classes
- Error
-
def
get_shift_tmax: (Int, Int)
Return the shift in categorical/treatment variable to make it start at zero as well as the maximum value after shifting.
Return the shift in categorical/treatment variable to make it start at zero as well as the maximum value after shifting. Must call 'dummyVars' first