The algorithm determines the controllable attribute as informative if the conditional mathematical expectation is in limits

and as non-informative, when value of conditional mathematical expectation are outside of these borders.

It is possible to observe the stochastic component of the load as consisting of expected deviation [zeta](t), which describes the

conditional mathematical expectation of the stochastic component and normally distributed non-correlated residual deviation (white noise) [xi](t).

It is possible to describe the stochastic component of the load by the expected deviation [zeta](t), which represents the conditional mathematical expectation of the stochastic component and normally distributed noncorrelated residual deviation (white noise) [xi](t) .

It is possible to found also short-term forecast deviation from conditional mathematical expectation of the load, which considers real progress of load in the recent past and also possible temperature influence.

It is possible to describe the stochastic component of the load by expected deviation [zeta]( ), which represents the

conditional mathematical expectation of the stochastic component and normally distributed noncorrelated residual deviation (white noise) [xi]t.

Here expected deviation of the load [zeta](t) describes the

conditional mathematical expectation of the stochastic component, which is also needed for calculating load short-term (more accurate) forecast.