1
$\begingroup$

I try to prove that the asymptotics of the recurrence $ T(n) = T(n-1) + \Theta (n) $ is $ T(n) = \Theta(n^2) $.

By $\Theta$, I mean tight bound from above and below.


I can write the equation like this: $$ T(n) = T(n-1) + \Theta (n) = T(n-1) + cn. $$

I'm having a hard time proving the lower bound with this method, do you have any idea how?

What I did for Upper bound is:

Assume: $$ T(n-1) \leq c(n-1)^2, \quad c \in R $$

Therefore: $$ T(n) \leq c(n-1)^2 + cn = cn^2 - 2cn + c + cn = cn^2 - cn + c \leq cn^2 $$

Therefore I get: $$ T(n) \leq O(n^2) $$

For $O$, by which I mean just an upper bound.

So, how do I prove a matching lower bound using this method?

$\endgroup$

1 Answer 1

1
$\begingroup$

Let $f(n) = \Theta(n)$. Thus there are $N,a,b>0$ such that for $n \geq N$, we have $an \leq f(n) \leq bn$.

Now suppose that $T(n) = T(n-1) + f(n)$, with some base case $T(0) = C$. Unrolling the sum, $$ T(n) = C + f(1) + \cdots + f(n). $$ Let $D = C + f(1) + \cdots + f(N-1)$. For $n \geq N$, $$ T(n) \leq D + b(N + \cdots + n) = D + b\frac{(n-N+1)(n+N)}{2} = O(n^2). $$ Similarly, $$ T(n) \geq D + a(N + \cdots + n) = D + a\frac{(n-N+1)(n+N)}{2} = \Omega(n^2). $$

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.