Lv 731,482 points

# Ian H

Favourite answers13%
Answers8,605
• ### A DIVERSION for your entertainment?

We know that the harmonic series diverges

H = 1/1 + 1/2 + 1/3 + 1/4 + 1/5 + ... = inf

The sum of reciprocals of square roots, namely

S = 1/√(1) + 1/√(2) + 1/√(3) + 1/√(4) + 1/√(5) + 1/√(6) + 1/√(7) + 1/√(8)

must also diverge, (because the denominators are all smaller).

But suppose we divide the inf sum S into two parts, like this

Let O be the inf sum with odd terms, O = 1/√(1) + 1/√(3) + 1/√(5) + 1/√(7)

and let E be the sum with even terms, E = 1/√(2) + 1/√(4) + 1/√(6) + 1/√(8)

We have arranged that O + E = S, term by term, but consider the series √(2)E

√(2)E = √(2)/√(2) + √(2)/√(4) + √(2)/√(6) + √(2)/√(8) + ...

√(2)E = 1/√(1) + 1/√(2) + 1/√(3) + 1/√(4) = S = E + O, so now we have

O = [√(2) – 1]E ~ 0.4142E

This result tells us that the odd sum is less than the even sum! But odd terms of the form 1/√(2n - 1) are larger than corresponding even terms 1/√(2n) which means that this is impossible. They both diverge of course, but which sum to n terms is larger? Jacob Bernoulli remarked on this apparent paradox.

Spoiler alert

>

>>>>>>>>

>

>

>

>

>

>

>

>

>

A quick calculation comparing 1000 terms resolves the issue.

Σ1/√(2n) ~ 43.6999

Σ1/√(2n - 1) ~ 44.2936

Although they both head off to infinity, comparable odd terms are always larger, so that series O is always slightly ahead at any intermediate stage.

The apparent result O ~ 0.4142E was just wrong.

But can you explain why?

1 AnswerMathematics3 weeks ago
• ### Series paradox question?

Consider the infinite series S

S = 1/√(1) + 1/√(2) + 1/√(3) + 1/√(4) + 1/√(5) + 1/√(6) + 1/√(7) + ......

Jacob Bernoulli knew that this sum of reciprocals of square roots, must

diverge since the denominators were all smaller than the harmonic series.

Let the inf sum of odd terms, O = 1/√(1) + 1/√(3) + 1/√(5) + 1/√(7) + .....

and the inf sum of even terms, E = 1/√(2) + 1/√(4) + 1/√(6) + 1/√(8) + ...

By inspection O + E = S, but we may also write √(2)E = S, so we have

O + E = √(2)E, or,

O = [√(2) – 1]E  ~ 0.4142E

Bernoulli remarked on the apparent paradox that the odd sum seems less than the even sum, but this is impossible because term by term odd terms are larger

Can you resolve the paradox with a clear explanation?

2 AnswersMathematics5 months ago
• ### Is anyone else getting “Certificate Error” warning when trying to log in here ? https://uk.answers.yahoo.com/dir/index/answer?sid=396545161?

Notices and Errors6 months ago
• ### What s the sum from 1 to infinity of the product cos(pi/n)*cos(2pi/n)*cos(3pi/n)*...*cos(npi/n) .... How do we calculate this?

Mathematics7 months ago
• ### Find the three integers a, b, c, such that are related by squares as:- a + b = x^2 b + c = y^2 a + c = z^2 a + b + c = w^2?

2 AnswersMathematics7 months ago
• ### Complex integral around unit circle at origin for 1/z is 2𝛑i, For 1/z^n it is zero as per Cauchy s first even though pole there. Why?

The function f(z) = 1/z, is

not analytic at the origin; it has a type of singularity there called a pole.

Accepting that result and the reason given, it seemed reasonable that

if f(z) = 1/z^n, around a unit circle, then ∮(1/z^n)dz would also NOT be zero.

But the result of calculation is that ∮(1/z^n)dz = 0, (unless n = 1)

4 AnswersMathematics1 year ago
• ### How to uplift an image of a diagram, (a triangle for example), for yahoo maths answers.?

2 AnswersYahoo Answers2 years ago
• ### Does anyone know why answers to posts often get this comment in red? Sorry, we were unable to proceed with your request. Please try again.?

3 AnswersSingles & Dating2 years ago
• ### COMPLEX Suppose that f(z) is analytic, with f(0) = 3 - 2i and f(1) =6 - 5i . Find f(1 + i) if Im f (z) = 6x(2y - 1)?

2 AnswersMathematics2 years ago
• ### COMPLEX UNSOLVED Suppose that f(z) is analytic, with f(0) = 3 - 2i and f(1) =6 - 5i . Find f(1 + i) if Im f'(z) = 6x(2y - 1)?

2 AnswersMathematics2 years ago
• ### COMPLEX FUNCTION QUESTION Suppose that f(z) is analytic, with f(0) = 3 - 2i and f(1) =6 - 5i . Find f(1 + i) if Im f'(z) = 6x(2y - 1) ?

1 AnswerMathematics2 years ago
• ### COMPLEX FUNCTION QUESTION Suppose that f(z) is analytic and f(0)=3 - 2i and f(1)=6 - 5i . Find f(1 + i) if Im f'(z) = 6x(2y - 1) ?

1 AnswerMathematics2 years ago
• ### Complex angle and value question Arg (z - i)/(z + i) = pi/4 (1) It is not obvious at the start that z + i is real, so how to solve (1) ?

4 AnswersMathematics2 years ago
• ### Prove (x + y + z)/3 ≥ (xyz)^(1/3) by ALGEBRA like adding 4xy to (x – y)^2 ≥ 0 to get AM–GM for 2 variables to get (x + y)/2 ≥ √(xy)?

3 AnswersMathematics3 years ago
• ### Other identities for sum of 3 squares itself a square? 1 ^2 + 2 ^2 + 2 ^2 = 3 ^2 is from x^2 + (x + 1)^2 + (x^2 + x)^2 = (x^2 + x + 1)^2?

2 AnswersMathematics3 years ago
• ### What is the UK prime ministers name ?

4 AnswersGovernment5 years ago
• ### Before computer fail got to level 7, as Ian H, with 37% best answers. New computer now, but how can status be regained ?. Regards - Ian H?

2 AnswersPolls & Surveys6 years ago
• ### CONICS: 2x^2 -6xy +2y^2 +43x +43y – 174 = 0, x^2 +y^2 +5x +5y – 30 = 0 Please recommend a free online software to plot both. Regards - Ian H?

5 AnswersMathematics6 years ago