MHB Upper and Lower bound of the recursive relation

  • Thread starter Thread starter evinda
  • Start date Start date
  • Tags Tags
    Bound Relation
Click For Summary
The discussion focuses on determining the asymptotic upper and lower bounds for the recursive relation T(n) = 5T(n/5) + n/ln(n). The master theorem is initially considered for application, but doubts arise regarding its suitability due to the nature of f(n) being asymptotically smaller than n. A substitution method is proposed, transforming the relation using m = log_5(n) to simplify the analysis. The resulting equation leads to a summation involving S(m), which requires further exploration to derive the bounds. The conversation highlights the challenges in applying the master theorem and the potential of substitution for solving the recursive relation.
evinda
Gold Member
MHB
Messages
3,741
Reaction score
0
Hello! (Wave)

I have to define an asymptotic upper and lower bound of the recursive relation $T(n)=5 T(\frac{n}{5})+\frac{n}{ \lg n}$.

I thought that I could use the master theorem,since the recursive relation is of the form $T(n)=aT(\frac{n}{b})+f(n)$

$$a=5 \geq 1 , b=5>1 , f(n)=\frac{n}{ \lg n}$$

$$f'(n)=\frac{ \lg n-1}{ \lg^2 n}>0 \Rightarrow \lg n >1 \Rightarrow n>2$$

So, $f(n)$ is asymptotically positive and increasing $\forall n>2$.

$$n^{\log_b a}=n^{\log_5 5}=n$$

We see that $f(n) < n$

$$f(n)=O(n^{ \log_b a- \epsilon})=O(n^{1- \epsilon})$$

But how can we find the $\epsilon$ ? (Thinking) Or can't we apply in this case the master theorem? (Worried)
 
Physics news on Phys.org
I think that the master theorem cannot be applied. (Shake)

But..we could do it with substitution,right? (Thinking)

We set $m= \log_5{n} \Rightarrow n=5^m$

Then,we have:

$$T(5^m)=5T(5^{m-1})+\frac{5^m}{m} \\ \Rightarrow \frac{T(5^m)}{5^m}=\frac{T(5^{m-1})}{5^{m-1}}+\frac{1}{m}$$

Let $S(m)=\frac{T(5^m)}{5^m}$

So:

$$S(m)=S(m-1)+\frac{1}{m} \\ S(m-1)=S(m-2)+\frac{1}{m-1} \\ \dots \\ \dots \\ S(2)=S(1)+\frac{1}{2} \\ + --------------- \\ \Rightarrow S(m)=S(1)+\left ( \frac{1}{2}+ \dots + \frac{1}{m}\right )$$

But,how could we continue? (Thinking)
 
Greetings, I am studying probability theory [non-measure theory] from a textbook. I stumbled to the topic stating that Cauchy Distribution has no moments. It was not proved, and I tried working it via direct calculation of the improper integral of E[X^n] for the case n=1. Anyhow, I wanted to generalize this without success. I stumbled upon this thread here: https://www.physicsforums.com/threads/how-to-prove-the-cauchy-distribution-has-no-moments.992416/ I really enjoyed the proof...

Similar threads

  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 17 ·
Replies
17
Views
2K
  • · Replies 2 ·
Replies
2
Views
1K
  • · Replies 3 ·
Replies
3
Views
5K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
13
Views
3K
  • · Replies 8 ·
Replies
8
Views
2K