Ram Murty 1/8 Partial summation formula and applications
Why Mathematics Stack Exchange recommends it
Mathematics Stack Exchange users cite it for the probabilistic way of thinking about divisibility and Euler's totient formula, which Murty presents around the 17th minute.
Recommended by 0 people across 0 different months, 2018 to 2024.
What it is
The first lecture in Ram Murty's analytic number theory course, recorded at Queen's University and re-uploaded with fixed audio. It covers the partial summation formula and its applications, and also uses probabilistic heuristics, such as the chance that a random integer is divisible by a prime.
What people said
Cites user572425 · 2018-10 · Mathematics Stack Exchange · on “Question about the probability of a number”
This is in reference to our video [this video] which is Ram Murty's video on Analytic Number Theory. On the 17th minute the professor says what is the probability that a random number is divisible by $p$ ? and he himself answers that the probability is $\frac{1}{p}$ ? However I dont understand how he found that? Am I missing something ?Is the answer so much easy?
Cites Luke Collins · 2020-04 · Mathematics Stack Exchange · on “probability that two primes $p_1$ and $p_2$ divide a positive integer”
@joriki I edited my answer. In number theory it's very common to talk about probabilities this way. For instance, the fact that the probability of $x$ being divisible by $p$ is $1/p$ leads to the fact that $\phi(x) = x\prod_{p\mid x}(1-\frac1p)$. Here [this video] is Ram Murty giving this example.
Cites aram · 2024-07 · Mathematics Stack Exchange · on “Probabilistic proof of $\varphi(n) = n\prod_{p \mid n} \left(1 - \frac{1}{p}\right)$”
In these video lectures [this video] Ram Murty gives a, I think, very intuitive to remember probabilistic proof of the following equality of the Euler $\varphi$ function: $$ \varphi(n) = n\prod_{p \mid n} \left(1 - \frac{1}{p}\right) $$ Proof: $$\dfrac{\varphi(n)}{n}$$ equals to the probability that a number is coprime with $n$ (ocurrences over total cases). On the other hand a number is coprime with $n$ if it doesn't divide any of the prime divisors of $n$ , the probability of $k \nmid p$ with $p \mid n$ is $$\dfrac{p-1}{p}$$ so because these are independent and we want the number to not divide any of the primes that divide $n$ we can multiply them to get the desired product. My issue: Although