1
$\begingroup$

I have three values, two of them are from $0 - 144$ and one is from $0 - 24$. I want to normalize these values and end up with a value from $0 - 1$ or $0 - 100$. I wanted to know if I can use the following equation to do this? And if so, how to go about it.

$$ z_i=\frac{x_i-\min(x)}{\max(x)-\min(x)} $$

where $x=(x_1,...,x_n)$ and $z_i$ is now your $i^{th}$ normalized data.

Applying this to my example, if I have three values to get an overall number from $0-1$ or $0-100$ do I have to do put each value individually though this equation and then add them or can I add the values together and use this as the max and do it this way? Also for min wouldn't this always be zero ?

$\endgroup$
1
$\begingroup$

If you want to scale your values with range $[\min,\max]$ to the new range $[a, b]$ the formula is:

$$ z_{i} = \frac{(b - a)(x_{i} - \min(x))}{\max(x) - \min(x)} + a $$

Apply this formula for each of your original values. Here is an example: Say you have three values $x_1, x_2, x_3$. The first two values have a range of $[0, 144]$ and the third has a range of $[0, 24]$. You want to transform all three to have a new range of $[0, 100]$. Let's say your values are: $x_1 = 133.43, x_2 = 51.65, x_3 = 6.91$. The corresponding transformations are thus:

\begin{align} z_1 &= \frac{(100 - 0)(133.43 - 0)}{144 - 0} + 0 = 92.66 \\ z_2 &= \frac{(100 - 0)(51.65 - 0)}{144 - 0} + 0 = 35.87 \\ z_3 &= \frac{(100 - 0)(6.91 - 0)}{24 - 0} + 0 = 28.79 \\ \end{align}

$\endgroup$
  • $\begingroup$ Thank you for your answer. To clarify, this has scaled the values to the data set I have correct. Also this is a bit of a different question, but if I add the three values outputted and put it over the total (eg. 144+144+24) would this represent the percentage of these values over the dataset or is it just an arbitrary number? $\endgroup$ – dmnte Dec 16 '17 at 14:26
  • $\begingroup$ @dmnte Yes, this has rescaled the three values. About the second question: I don't think adding them is useful in this context. $\endgroup$ – COOLSerdash Dec 16 '17 at 14:30

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.