Changes between Version 6 and Version 7 of Ticket #635


Ignore:
Timestamp:
Sep 2, 2009, 8:37:52 AM (17 years ago)
Author:
Víctor de Buen Remiro
Comment:

Legend:

Unmodified
Added
Removed
Modified

  • Ticket #635 – Description

    v6 v7  
    33A tipical situation is that an input or output is non stationary, so scalar truncated normal distributions are no too apropriate.
    44
    5 For example, if the original series requires a regular difference to be stationary, then user could write something like this
     5Then, it should be possible to define an ARIMA model for the residuals of a special node {{{missingResidual}}} with linear equations of regression:
    66
    77{{{
    8   knownValue[n-1] = missingResidual[n] + missingVariable[n]
     8  knownValue[n] = missingResidual[n] //If n-th datum is known
    99}}}
    1010
    11 if n-th datum is unknown but (n-1)-th is known, or
    12 
    1311{{{
    14   0 = missingResidual[n] + missingVariable[n] - missingVariable[n-1]
     12  0 = missingResidual[n] - missingVariable[n] //If n-th datum is unknown
    1513}}}
    1614
    17 if both n-th and (n-1)-th data are unknown.