The master method is commonly used to evaluate [[algorithmic efficiency]] for algorithms that use [[base/Algorithms/recursion|recursion]].
To use the master method, first express the [[recurrence]] of the algorithm in the form
$T(n) =
\begin{cases}
\Theta(1) \text{ if } n\le 1 \\
a T \Big (\frac{n}{b} \Big) + \Theta(n^c)
\end{cases}
$
Next simply lookup the correct runtime from the table below where $\epsilon = log_b a$.
| Case | Runtime |
| -------------- | ----------------------------- |
| $\epsilon > c$ | $\Theta(n^{\epsilon})$ |
| $\epsilon = c$ | $\Theta(n^{\epsilon} \lg n)$) |
| $\epsilon < c$ | $\Theta(n^c)$ |