I'm reading the book "Fault-Diagnosis Systems" by Isermann in the par. 9.2.1a. The author explains how to estimate the parameter of a linear difference equation using Least Squares. We start with a standard ordinary linear equation with an error term:

enter image description here

the equation is called 9.15. Then,

enter image description here

The first thing i don't understand is the syntax used by the author. What's the meaning of the | inside $\hat{y}(k|k-1)$ ?

ps: I'm sorry if what i ask is trivial, but when i have a kind of doubt like this, i don't know how to solve it if not asking it in this site...

  • $\begingroup$ Hi: that notation denotes the prediction ( also referred to as estimate ) of $y(k)$ at time $(k-1)$. $\endgroup$ – mark leeds Jan 26 '19 at 13:45
  • $\begingroup$ Is the conditional probability ? $\endgroup$ – Jhdoe Jan 26 '19 at 14:15
  • $\begingroup$ Hi: No. Assume the model is estimated which means that the coefficients are estimated. Then, that notation denotes the prediction of the next value in the series at time $k$, given that one is at time $k-1$. Maybe check out a book or a discussion on arima models if this is not clear. $\endgroup$ – mark leeds Jan 27 '19 at 4:01
  • $\begingroup$ So it's like the author is saying : "suppose we use a model to predict y. Than we get something like this" .... :S ? $\endgroup$ – Jhdoe Jan 27 '19 at 10:23
  • $\begingroup$ yes. but y ONE step ahead. time matters a lot. $\endgroup$ – mark leeds Jan 27 '19 at 19:09

The notation $ \hat{y} \left( k \mid k - 1 \right) $ usually means this is an estimated value of $ y \left( k \right) $ given all the available data up to time index $ k - 1 $.
So generally speaking, this is a prediction of one step in time of the data.

The case above also suggests linear estimation.
Namely, $ \hat{y} \left( k \right) $ is built using linear combination of all the given data.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.