Mean Absolute Error (meanabs_err)
Contents
Mean Absolute Error. Mean Absolute for Linear Loss scenarios function. Calculated by averaging over scenarios the absolute values of losses .
Syntax
meanabs_err(matrix) |
short call |
meanabs_err_name(matrix) |
call with optional name |
Parameters
matrix is a Matrix of Scenarios:
where the header row contains names of variables (except scenario_probability, and scenario_benchmark). Other rows contain numerical data. The scenario_probability, and scenario_benchmark columns are optional.
Output
When function Mean Absolute Error is used in optimization or calculation problems PSG automatically calculates and includes in the solution report two outputs:
pseudo_R2_function_name |
|
contributions(function_name) |
Mathematical Definition
Mean Absolute Error function is calculated as follows
,
where:
E denotes the expectation sign;
random vector 
has components 
and J vector scenarios, 
,
random value 
, which is the i-th component of the random vector, 
, has J discrete scenarios 
,
is probability of the scenario 
.
is Loss Function (See section Loss and Gain Functions)
is an argument of Mean Absolute Error function.
Example
See also
Mean Absolute Error Normal Independent, Mean Absolute Error Normal Dependent, Mean Square Error, Mean Square Error Normal Independent , Mean Square Error Normal Dependent, Root Mean Squared Error, Root Mean Squared Error Normal Independent, Root Mean Squared Error Normal Dependent, Koenker and Basset Error, Rockafellar Error