I have a sum of real numbers $A_1 + A_2 + \cdots + A_N = A$ that add up to a known number $A$. All of the $A_1,\ldots,A_N$ are known as well.

Is there a way of scaling the $A_1,\ldots,A_N$ so that the sum of the numbers add up to another known real number $B$ instead of $A$?

So what I am searching for are $\gamma_1,\ldots,\gamma_N$ such that:

$$A_1\gamma_1 + A_2\gamma_2 + \cdots + A_N\gamma_N = B$$

What assumption can I make to ensure uniqueness of the $\gamma_1,\ldots,\gamma_N$? Is it possible to find $\gamma_1,\ldots,\gamma_N$?


You can make all the $\gamma$'s$= \frac BA$. There are many other solutions-you can make the first $N-1$ anything you want and you will have a linear equation for the last one.

  • 1
    $\begingroup$ As long as $A_N\neq0$, anyway. $\endgroup$ – Cameron Buie Nov 17 '12 at 17:18
  • $\begingroup$ @RossMillikan: Thanks, Ross; I really like the $\gamma = \frac BA$ solution. What are the conceptual implications of making $\gamma = \frac BA$? What type of scaling is this? $\endgroup$ – Nicholas Kinar Nov 17 '12 at 17:37
  • $\begingroup$ It is just a linear scaling. I'm not sure what you mean by conceptual implications. It is like you graded on a scale of 0 to 5 and now want 0 to 100. You multiply everything by 20. $\endgroup$ – Ross Millikan Nov 17 '12 at 17:46
  • $\begingroup$ @RossMillikan: Thanks, Ross. That's exactly what I was looking for. $\endgroup$ – Nicholas Kinar Nov 17 '12 at 17:52

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.