Let's make the signs alternate

Algebra Level 2

What number goes into the blank below?

1 + 1 2 + 1 4 + 1 8 + 1 16 + = ( 1 1 2 + 1 4 1 8 + 1 16 ) × _______ {\color{#3D99F6}1} + {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac14} + {\color{#D61F06}\dfrac18} + {\color{#3D99F6}\dfrac1{16}} + \cdots =\left({\color{#3D99F6}1} - {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac14} - {\color{#D61F06}\dfrac18} + {\color{#3D99F6}\dfrac1{16}} - \cdots \right) \times \text{\_\_\_\_\_\_\_}


The answer is 3.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

9 solutions

Calvin Lin Staff
Jul 19, 2017

Let A = 1 + 1 4 + 1 16 + A = 1 + \frac{1}{4} + \frac{ 1}{16} + \ldots and
let B = 1 2 + 1 8 + 1 32 + B = \frac{1}{2} + \frac{ 1}{8} + \frac{1}{32} + \ldots .

We see that 2 B = A 2B = A . The question asks:

A + B = __ ( A B ) A + B = \text{\_\_} ( A - B)

Substituting in B = A 2 B = \frac{A}{2} , we get 3 2 A = __ ( 1 2 A ) \frac{3}{2} A = \text{\_\_} ( \frac{1}{2} A) .

Hence, the answer is 3.


Note: What assumption does this solution make?

Moderator note:

In the comments Calvin suggests trying the same procedure with

1 + 1 2 + 1 3 + 1 4 + 1 5 + = _______ × ( 1 1 2 + 1 3 1 4 + 1 5 ) {\color{#3D99F6}1} + {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac13} + {\color{#D61F06}\dfrac14} + {\color{#3D99F6}\dfrac1{5}} + \cdots = \text{\_\_\_\_\_\_\_} \times \left({\color{#3D99F6}1} - {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac13} - {\color{#D61F06}\dfrac14} + {\color{#3D99F6}\dfrac1{5}} - \cdots \right)

to see how it is flawed when A A goes to infinity or is undefined.

Using the same process:

Let A = 1 + 1 2 + 1 3 + A = 1 + \frac{1}{2} + \frac{ 1}{3} + \ldots and
let B = 1 2 + 1 4 + 1 6 + B = \frac{1}{2} + \frac{ 1}{4} + \frac{1}{6} + \ldots .

Again, we see that 2 B = A 2B = A , implying (via the same steps as before) 3 2 A = __ ( 1 2 A ) \frac{3}{2} A = \text{\_\_} ( \frac{1}{2} A) . This gets the same answer of 3 as the main problem.

However, the series on the left hand side of the equation is the harmonic series which is known to diverge to infinity and the series on the right hand side of the equation is the alternating harmonic series which is known to converge to ln 2 0.693. \ln 2 \approx 0.693. Clearly, \infty does not even approximately equal 3 × 0.693. 3 \times 0.693 .

Is the assumption that both series converge and we can do finite arithmetic with them?

Zach Abueg - 3 years, 10 months ago

Log in to reply

Right. If A = A = \infty or A A is undefined, then we cannot conclude what the expression is equal to. Most people forget about that possibility because they solve for the value over the real numbers.

As an example, try solving for:

1 + 1 2 + 1 3 + 1 4 + 1 5 + = _______ × ( 1 1 2 + 1 3 1 4 + 1 5 ) {\color{#3D99F6}1} + {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac13} + {\color{#D61F06}\dfrac14} + {\color{#3D99F6}\dfrac1{5}} + \cdots = \text{\_\_\_\_\_\_\_} \times \left({\color{#3D99F6}1} - {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac13} - {\color{#D61F06}\dfrac14} + {\color{#3D99F6}\dfrac1{5}} - \cdots \right)

using the same approach.

Calvin Lin Staff - 3 years, 10 months ago

Log in to reply

There are 2 ways we could approach it with methods similar to yours. We could either select A and B A \text{ and } B such that A = 2 B A = 2B , or select it such that we can write the original problem as ( A + B ) = __ ( A B ) (A + B) = \text{\_\_}(A - B)

For Method One: let A = 1 + 1 2 + 1 3 + 1 4 + 1 5 + A = 1 + \frac 12 + \frac 13 + \frac 14 + \frac 15 + \ldots and let B = 1 2 + 1 4 + 1 6 + 1 8 + 1 10 + B = \frac 12 + \frac 14 + \frac 16 + \frac 18 + \frac 1{10} + \ldots Indeed this does give us A = 2 B A = 2B but, it also gives us A = __ ( A 2 B ) A = \text{\_\_}(A-2B) Substituting A A for 2 B 2B gives us A = __ ( 0 ) A = \text{\_\_}(0) This would force A = 0 A = 0 , which isn't true since the harmonic series diverges to infinity. It claims that the alternating harmonic series is also equal to 0, though that seems to be a result of its conditional, rather than absolute, convergence.

For Method Two: let A = 1 + 1 3 + 1 5 + A = 1 + \frac 13 + \frac 15 + \ldots and let B = 1 2 + 1 4 + 1 6 + B = \frac 12 + \frac 14 + \frac 16 + \ldots This allows us to rewrite your orignal problem as ( A + B ) = __ ( A B ) (A + B) = \text{\_\_}(A - B) , but I see no relation that we can make between A and B to get further into the problem

Jacob Huebner - 3 years, 10 months ago

You show A as the sun off all terms in the series. Shouldn't it be alternating terms? Just adding the first two alternating terms shows that 2B≠A

norbert turek - 3 years, 10 months ago

Log in to reply

Read carefully. A is the sum of the red terms, B is the sum of the blue terms.

Each individual term of A is twice the corresponding individual term of B, "hence" 2 B = A 2B = A .

Calvin Lin Staff - 3 years, 10 months ago

3/2-a/2=1/3 MCB

Michael Butler - 3 years, 10 months ago

Nice solution

AMAN KUMAR - 3 years, 10 months ago

I don't see how it's an 'assumption' that the geometric series converges. Or that the sum of two series, each of which converges absolutely, also converges.

Richard Desper - 3 years, 10 months ago

Log in to reply

The point is that "A is finite" is an assumption, up to the point where we say "A is finite because it is a geometric series that converges". Yes, that observation might be obvious to you, but it needs to be stated.

IE If you look at the other example, where the assumption of "A is finite" is false, then you see how the argument breaks down.

Calvin Lin Staff - 3 years, 10 months ago

Regarding the Challenge Master note: You say "using the same process", but you didn't do that. Calvin's process placed each component of the sum into either A or B. In your process, every component in B is also in A.

Jerry Barrington - 3 years, 10 months ago

@Calvin to be clear, in the second example , Challenge Master notes " Using the same process: let A= 1+ 1/2 + 1/3 + ..." Those are not the alternating terms. Your example is correct.

norbert turek - 3 years, 10 months ago

Log in to reply

Ah, I see that the process is consistent.

norbert turek - 3 years, 10 months ago

Yeah, I didn't scroll over on my phone. Thought the first blank was the entire question. Was amazed it wasn't 2.

Steve Powersuit - 3 years, 10 months ago

Log in to reply

Thanks. Can you confirm if this is on your phone browser, or on the app? I have filed a bug report and we will look into it.

In the meantime, I have updated the question so that if the problem is cut off, then it cannot be answered at all (which is a slight improvement over answering the wrong problem).

Calvin Lin Staff - 3 years, 10 months ago

hi Calvin. on my Samsung browser, the actual problem does not appear in totality until a wrong answer is given.

Jeannine Myer - 3 years, 10 months ago

Log in to reply

Thanks. I have filed a bug report and we will look into it.

In the meantime, I have updated the question so that if the problem is cut off, then it cannot be answered at all (which is a slight improvement over answering the wrong problem).

Calvin Lin Staff - 3 years, 10 months ago
Marta Reece
Jul 18, 2017

Using the formula for a sum of geometric series S = a 1 r S=\frac a{1-r} with a = 1 a=1 and r = 1 2 r=\frac12 on the left and a = 1 a=1 and r = 1 2 r=-\frac12 on the right.

1 1 1 2 = x × 1 1 + 1 2 \dfrac1{1-\frac12}=x\times\dfrac1{1+\frac12}

2 = 2 3 x 2=\frac23x

x = 3 x=\boxed3

Irrational algebra

Michael Butler - 3 years, 10 months ago

Log in to reply

Nothing irrational about this.

Pi Han Goh - 3 years, 10 months ago

Yes, for clarity, you should state that this is a convergent geometric series, otherwise, we can obtain the Grandi's series: 1 1 + 1 1 + = 1 2 1 - 1 + 1 - 1 + \cdots = \frac12 .

Bonus: In order for the geometric series to converge, its common ratio must satisfy r < 1 |r| < 1 . Can you explain why?

Pi Han Goh - 3 years, 10 months ago
Nicolai Hinsch
Jul 24, 2017

The sum of the Progression of 1 2 n \frac{1}{2^n} is equal to two. I visualized this with two squares. The second sum diverges to 2 3 \frac{2}{3} . The convergence is simple to proof when you set brackets. (1- 1 2 \frac{1}{2} ) + ( 1 4 \frac{1}{4} - 1 8 \frac{1}{8} ) + ... is equal to the sum of 1 2 2 n 1 \frac{1}{2^{2*n-1}} for n > 0.

This series converges to 2 3 \frac{2}{3} , since the difference between the sum is always for 1 2 2 n 1 \frac{1}{2^{2*n-1}} equal to 1 ( 2 2 n 1 3 ) \frac{1}{(2^{2*n-1}*3)} . 2 3 \frac{2}{3} * 3 = 2

You can see, that the alternate series is 1 3 \frac{1}{3} of the geometric series good in the visualization, I choosed for this.

FYI To denote "a single term" in LaTeX, use { } around all of the variables. I've edited the first fraction for you as an example.

Calvin Lin Staff - 3 years, 10 months ago

That's a really creative way to explain this!! I definitely understood this better than the other explanation. ('better' doesn't mean I understood it completely; other than the picture, the rest of the explanation was like space talk to me. but still it helped a lot in visualizing it)

Young Ju Yoon - 3 years, 10 months ago

Log in to reply

Thanks a lot :-) I am glad that the visualization works for you! :-)

Nicolai Hinsch - 3 years, 10 months ago
Michael Stolnicki
Jul 24, 2017

Reorganize the sequence

Left side:

= (1+ 1 2 \frac{1}{2} ) + ( 1 4 \frac{1}{4} + 1 8 \frac{1}{8} ) + ( 1 16 \frac{1}{16} + 1 32 \frac{1}{32} ) + ...

= 3 2 \frac{3}{2} + 3 8 \frac{3}{8} + 3 16 \frac{3}{16} + ...

Right side:

= (1- 1 2 \frac{1}{2} ) + ( 1 4 \frac{1}{4} - 1 8 \frac{1}{8} ) + ( 1 16 \frac{1}{16} - 1 32 \frac{1}{32} ) + ...

= 1 2 \frac{1}{2} + 1 8 \frac{1}{8} + 1 16 \frac{1}{16} + ...

Hence, the answer is 3.

Hmmm, while you certain has the right idea (of grouping the numbers in pairs), using your logic, we can conclude that

1 1 + 1 1 + 1 1 + = ( 1 1 ) + ( 1 1 ) + ( 1 1 ) + = 0 1 - 1 + 1 - 1 + 1 - 1 + \cdots = (1-1) + (1-1) + (1-1) + \cdots = 0 and 1 1 + 1 1 + 1 1 + = 1 ( 1 1 ) ( 1 1 ) = 1 0 0 = 1 1 - 1 + 1 - 1 + 1 - 1 + \cdots = 1 - (1 - 1) - (1-1) - \cdots = 1 - 0 - 0 - \cdots = 1 .

So what is the value of 1 1 + 1 1 + 1 1 + 1 - 1 + 1 - 1 + 1 - 1 + \cdots ? Is it 0 or 1?

Pi Han Goh - 3 years, 10 months ago
Rocco Dalto
Jul 23, 2017

Since the given series n = 0 ( 1 ) n 1 2 n \sum_{n = 0}^{\infty} (-1)^n \dfrac{1}{2^n} converges absolutely we can rearrange the terms so that:

n = 0 1 4 n = 4 3 \sum_{n = 0}^{\infty} \dfrac{1}{4^n} = \dfrac{4}{3}

and,

1 2 n = 0 1 4 n = 2 3 -\dfrac{1}{2} \sum_{n = 0}^{\infty} \dfrac{1}{4^n} = -\dfrac{2}{3}

n = 0 ( 1 ) n 1 2 n = 2 3 \implies \sum_{n = 0}^{\infty} (-1)^n \dfrac{1}{2^n} = \dfrac{2}{3}

and n = 0 1 2 n = 2 2 = 2 3 j j = 3. \sum_{n = 0}^{\infty} \dfrac{1}{2^n} = 2 \implies 2 = \dfrac{2}{3} j \implies j = 3.

Joshua Powles
Jul 28, 2017

An alternative way to tackle this question is by finding the limits of the geometric series. These series are famous enough that some may know them by heart, but here is the method:

The sum of the first n n terms with first term a a and common ratio r 1 r \neq 1 is:

S n = a ( 1 r n ) 1 r S_n=\frac {a(1-r^n)}{1-r}

Now we take the limit as n n \rightarrow \infty

Provided that 1 r < 1 -1 \leq r < 1 , the limit simplifies to a 1 r \frac {a}{1-r}

Now we simply substitute

The LHS: ( 1 ) 1 ( 1 / 2 ) = 2 \frac {(1)}{1-(1/2)}=2

The RHS series: ( 1 ) 1 ( 1 / 2 ) = 2 / 3 \frac {(1)}{1-(-1/2)}=2/3

The missing value is x = 2 2 / 3 = 3 x=\frac {2}{2/3}=3

That's how I did it. I ignored the picture.

Richard Desper - 3 years, 10 months ago

Visualize: Draw a line (corner to corner) of right hand square, separating the square into two equal parts. Taking one so divided area as "A," it would take twice A (integer 2) to equal the equivalent blue square. But red areas are greater than A, so the integer has to be greater than 2. If the integer were 4, then the entire right hand square would be red. So the integer is 3.

John Wehrung - 3 years, 10 months ago
Akshay Gupta
Jul 26, 2017

1 + 1 2 + 1 4 + 1 8 + 1 16 + = _______ × ( 1 1 2 + 1 4 1 8 + 1 16 ) {\color{#3D99F6}1} + {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac14} + {\color{#D61F06}\dfrac18} +{\color{#3D99F6}\dfrac1{16}} + \cdots = \text{\_\_\_\_\_\_\_} \times \left({\color{#3D99F6}1} - {\color{#D61F06}\dfrac12} + {\color{#3D99F6}\dfrac14} - {\color{#D61F06}\dfrac18} + {\color{#3D99F6}\dfrac1{16}} - \cdots \right)

Let sum of G.P in L.H.S. = S1,

sum of G.P. in R.H.S. = S2

and blank space that we have to find be A

S1 = a 1 r \frac{a}{1-r} = 1 1 ( 1 2 ) \frac{1}{1-(\frac{1}{2})} = 2

S2 = a 1 r \frac{a}{1-r} = 1 1 ( 1 2 ) \frac{1}{1-(\frac{-1}{2})} = 2 3 \frac{2}{3}

S1 = A * S2

A = S 1 S 2 \frac{S1}{S2}

= 2 ( 2 3 ) \frac{2}{(\frac{2}{3})}

A = 3

Betty BellaItalia
Jul 29, 2017

Robert DeLisle
Jul 29, 2017

Considering the terms in pairs we see that each partial sum of the positive term series, taken in pairs, is three times the partial sum of the corresponding pairs of the alternating term series. Since we know that both series converge this is sufficient to get the answer 3.

That is, since

k = 0 2 n 1 2 k = 3 k = 0 2 n 1 ( 1 ) k 2 k \sum_{k=0}^{2n-1} 2^{-k} = 3 \sum_{k=0}^{2n-1} (-1)^k2^{-k} for all n > 0 and both series converge (sum of remaining terms can be made arbitrarily small) we can conclude that

k = 0 2 k = 3 k = 0 ( 1 ) k 2 k \sum_{k=0}^{\infty} 2^{-k} = 3 \sum_{k=0}^{\infty} (-1)^k2^{-k}

This approach is very similar to the approximation solution to 100 Day Summer Challenge Day 58 with nested squares instead of missing corners as in this problem. The ultimate fraction becomes apparent early on in a structural way in both.

One might also notice that both are geometric series that can be computed directly via 1 1 r \frac {1} {1-r} with r = 1 2 \frac{1}{2} and r = 1 2 \frac{-1}{2} respectively to get 2 and 2/3 for sums and the answer 3.

Hmmm, why must k = 0 2 n 1 2 k = 3 k = 0 2 n 1 ( 1 ) k 2 k \sum_{k=0}^{2n-1} 2^{-k} = 3 \sum_{k=0}^{2n-1} (-1)^k2^{-k} implies k = 0 2 k = 3 k = 0 ( 1 ) k 2 k \sum_{k=0}^{\infty} 2^{-k} = 3 \sum_{k=0}^{\infty} (-1)^k2^{-k} ? Does it work for all n>0? Why?

Pi Han Goh - 3 years, 10 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...