LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   Chapter 3 - The Linear Model (http://book.caltech.edu/bookforum/forumdisplay.php?f=110)
-   -   invalid points in Z transform (http://book.caltech.edu/bookforum/showthread.php?t=4670)

 pouramini 05-05-2016 08:05 AM

invalid points in Z transform

The book says

Quote:
 " In general, some points in the Z space may not be valid transforms of any x E X, and multiple points in X may be transformed to the same z E Z, depending on the nonlinear transform "
How a point in Z can be not a valid transform of any x?

I suppose any z is a map of a x, not!?

 ntvy95 05-05-2016 10:32 AM

Re: invalid points in Z transform

Quote:
 Originally Posted by pouramini (Post 12335) The book says How a point in Z can be not a valid transform of any x? I suppose any x will be mapped to a z! not!?
Here is my understanding:

Any should be mapped to a , but not any can be mapped to a : In other words, nonlinear transform may not be an onto function.

For example the nonlinear transform (given in the book), if or then there is no can be mapped to such because there is no such that and no such that .

Hope this helps.

 pouramini 05-06-2016 12:03 AM

Re: invalid points in Z transform

Thank you, yes that seems helpful if we regard all points in Z space, but I thought it speaks about the points in Z which are the mapping of a point in data set D.

 ntvy95 05-06-2016 08:20 AM

Re: invalid points in Z transform

Quote:
 Originally Posted by pouramini (Post 12338) Thank you, yes that seems helpful if we regard all points in Z space, but I thought it speaks about the points in Z which are the mapping of a point in data set D.
Hm, in my understanding, the quote "some points in the Z space may not be valid transforms of any x E X" regards all the points in the Z space. For example: If , then the point in the Z space cannot be a valid transform of any .

 All times are GMT -7. The time now is 08:55 PM.