http://www.mathworks.com/matlabcentral/newsreader/view_thread/325824
MATLAB Central Newsreader  A basic question
Feed for thread: A basic question
enus
©19942015 by MathWorks, Inc.
webmaster@mathworks.com
MATLAB Central Newsreader
http://blogs.law.harvard.edu/tech/rss
60
MathWorks
http://www.mathworks.com/images/membrane_icon.gif

Wed, 16 Jan 2013 16:00:09 +0000
A basic question
http://www.mathworks.com/matlabcentral/newsreader/view_thread/325824#895573
Jerry
Hello,<br>
<br>
I used this model [y = D+(AD)/(1+(x/C)^B)] and predicted all parameters using a vector of 'x' and 'y'. For a specific value of 'y', I would like to calculate a value for the 'x' using this model [x =((((D+(AD))/y)1)^1/B)*C] derived from the original model. But the calculated 'x' is not what it should be. I am assuming there is something wrong with this model [x =((((D+(AD))/y)1)^1/B)*C]. May someone give me a comment?<br>
<br>
Thanks,<br>
Jerry

Wed, 16 Jan 2013 16:32:08 +0000
Re: A basic question
http://www.mathworks.com/matlabcentral/newsreader/view_thread/325824#895575
someone
"Jerry " <jerrycholo@gmail.com> wrote in message <kd6iq9$mmv$1@newscl01ah.mathworks.com>...<br>
> Hello,<br>
> <br>
> I used this model [y = D+(AD)/(1+(x/C)^B)] and predicted all parameters using a vector of 'x' and 'y'. For a specific value of 'y', I would like to calculate a value for the 'x' using this model [x =((((D+(AD))/y)1)^1/B)*C] derived from the original model. But the calculated 'x' is not what it should be. I am assuming there is something wrong with this model [x =((((D+(AD))/y)1)^1/B)*C]. May someone give me a comment?<br>
> <br>
> Thanks,<br>
> Jerry <br>
<br>
Based on what little information provided<br>
(I assume A,B,C, & D are scalars), I suggest two things:<br>
<br>
1. Put parentheses around 1/B, i.e.<br>
<br>
x =((((D+(AD))/y)1)^(1/B))*C <br>
<br>
2. Try using elementbyelement operators:<br>
<br>
x =((((D+(AD))./y)1).^(1/B))*C <br>
<br>
See if those help.

Wed, 16 Jan 2013 18:27:10 +0000
Re: A basic question
http://www.mathworks.com/matlabcentral/newsreader/view_thread/325824#895582
Roger Stafford
"Jerry " <jerrycholo@gmail.com> wrote in message <kd6iq9$mmv$1@newscl01ah.mathworks.com>...<br>
> I used this model [y = D+(AD)/(1+(x/C)^B)] and predicted all parameters using a vector of 'x' and 'y'. For a specific value of 'y', I would like to calculate a value for the 'x' using this model [x =((((D+(AD))/y)1)^1/B)*C] derived from the original model. ....<br>
         <br>
There is an algebra error in your inverse. It should be<br>
<br>
x = C*((Ay)/(yD))^(1/B)<br>
<br>
Your expression with just 'y' in the denominator cannot be correct.<br>
<br>
A note of warning. If 1/B is a fractional power, you may get values of x which are not equal to the original x used to generate the yvalue. Fractional powers can have multiple results. For example (1)^(1/4) has four different results in the complex plane: (1+i)/sqrt(2), (1i)/sqrt(2), (1+i)/sqrt(2), and (1i)/sqrt(2).<br>
<br>
Roger Stafford