Don't go all mediant on me!

Algebra Level 1

I know that just adding numerators and denominators doesn’t work. For example,

1 2 + 2 3 1 + 2 2 + 3 . \frac{1}{2}+\frac{2}{3} \ne \frac{1+2}{2+3}.

Curious if this kind of addition ever works, I look for positive integers a , b , c , d a,b,c,d such that

a b + c d = a + c b + d . \dfrac ab + \dfrac cd = \dfrac{a+c}{b+d}.

After some searching, I conclude that such numbers don't exist. Am I right?

No Yes

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.

10 solutions

Mikhail Komin
Oct 7, 2018

a + c b + d \dfrac{a+c}{b+d} = a b + d \dfrac{a}{b+d} + c b + d \dfrac{c}{b+d}

If a,b,c,d are positive integers then a b \dfrac{a}{b} > a b + d \dfrac{a}{b+d} and c d \dfrac{c}{d} > c b + d \dfrac{c}{b+d} .

This implies

a b \dfrac{a}{b} + c d \dfrac{c}{d} > a b + d \dfrac{a}{b+d} + c b + d \dfrac{c}{b+d}

and

a b \dfrac{a}{b} + c d \dfrac{c}{d} > a + c b + d \dfrac{a+c}{b+d}

Moderator note:

In fact, a b > c d \frac{a}{b} > \frac {c}{d} , then a b > a + c b + d > c d \frac{ a}{b} > \frac{ a + c } { b + d } > \frac{ c}{d} . (This requires that the variables are positive)

This follows by thinking of the fractions as percentages (esp in probability). When we combine two experiments, we intuitively know that the success probability is between the original 2.
To prove this inequality, we can cross multiply the terms to compare.

Yeah this is a better explanation.

Jeet Dutta - 2 years, 8 months ago

Wow, cool explain!!! :D:D:D

Qiao Qiao - 2 years, 8 months ago

What if we use (1/2)+(2/1)=(1+2)/(2+1)

Anil Yadav - 2 years, 8 months ago

Log in to reply

1/2+2/1=2.5

(1+2)/(2+1)=1

2.5 != 1

Mikhail Komin - 2 years, 8 months ago

I believe that in order for this explanation to work, a=/=b=/=c=/=d should have been given?

(=/= means is not)

Sofie Weggeman - 2 years, 8 months ago

Log in to reply

no, this is not necessary. You can check a=b=c=d=1, it is still works.

Mikhail Komin - 2 years, 8 months ago

1/2+1/2=2/4 LOL

Bahriddin Abdiev - 2 years, 8 months ago

Log in to reply

If a,b, c,d ,k are integers and a/b=c/d=k,then (a+c)/(b+d)=k

Lee Yang - 2 years, 7 months ago

Uh, no it doesn't...

Richard Zhu - 2 years, 7 months ago

It does work if a=b=c=d. The problem should explicitly state that this case is excluded. The fact that b=c in the case presented is misleading.

Gabriel Chacón - 2 years, 8 months ago

Log in to reply

If a,b, c,d ,k are integers and a/b=c/d=k,then (a+c)/(b+d)=k

Lee Yang - 2 years, 7 months ago

Hey if a=-c and b=d since integers can be negative

Prashanth Dasari - 2 years, 7 months ago

Log in to reply

If a,b, c,d ,k are integers and a/b=c/d=k,then (a+c)/(b+d)=k

Lee Yang - 2 years, 7 months ago

But we just discuss about positive intgers on this question so we cant use a=-c

Ghally Arrahman - 2 years, 7 months ago

If a,b, c,d ,k are integers and a/b=c/d=k,then (a+c)/(b+d)=k

Lee Yang - 2 years, 7 months ago

What about 1/2 + 1/2 = (1+1)/(2+2)

oops.

Not equal.

Kermit Rose - 2 years, 7 months ago

a b + c d = a + c b + d a d + b c b d = a + c b + d ( a d + b c ) ( b + d ) = ( a + c ) b d a d b + a d 2 + b 2 c + b c d = a b d + c b d a d 2 + b 2 c = 0 \dfrac{a}{b} + \dfrac{c}{d} = \dfrac{a + c}{b + d} \Longrightarrow \dfrac{ad + bc}{bd} = \dfrac{a + c}{b + d} \Longrightarrow (ad + bc)(b + d) = (a + c)bd \Longrightarrow adb + ad^{2} + b^{2}c + bcd = abd + cbd \Longrightarrow ad^{2} +b^{2}c = 0 .

But as a , b , c , d > 0 a,b,c,d \gt 0 we can only have a d 2 + b 2 c > 0 ad^{2} + b^{2}c \gt 0 , in which case the answer is Yes \boxed{\text{Yes}} , there are no positive integer solutions to the given equation.

Note: We can have a solution if precisely one of a , c a,c is negative, e.g., ( a , b , c , d ) = ( 2 , 5 , 8 , 10 ) (a,b,c,d) = (2,5,-8,10) or ( 2 , 3 , 8 , 6 ) (2,3,-8,6) .

About the note, there are also infinitely many solutions as long as you pick a = c = 0 and b, d, b+d != 0.

Joël Ganesh - 2 years, 8 months ago

Log in to reply

Not just b+d != 0 but b != 0 and d != 0 as well. Otherwise the initial fractions would be invalid.

Jochen Morent - 2 years, 8 months ago

What if a and/or c are 0?

Δημήτρης Μενούνος - 2 years, 8 months ago

Log in to reply

The question specifies that a , c a,c , along with b , d b,d , are positive integers, so this isn't a concern here. But if we remove that condition, then if either a , c a,c is zero then for the equation to hold the other must be zero as well. The equation will hold also if a = c a = -c and b = d b = d .

Brian Charlesworth - 2 years, 8 months ago

What if a,b,c,d are all equal to 1?

Danning Zhan - 2 years, 8 months ago

Log in to reply

This wouldn't work, since a b + c d = 1 1 + 1 1 = 2 \dfrac{a}{b} + \dfrac{c}{d} = \dfrac{1}{1} + \dfrac{1}{1} = 2 and a + b c + d = 1 + 1 1 + 1 = 1 \dfrac{a + b}{c + d} = \dfrac{1 + 1}{1 + 1} = 1 .

It would work if ( a , b , c , d ) = ( 1 , 1 , 1 , 1 ) (a,b,c,d) = (1,1,-1,1) , but for this problem we require that all of a , b , c , d a,b,c,d are positive.

Brian Charlesworth - 2 years, 8 months ago
Peter Macgregor
Oct 8, 2018

Let us assume that a b + c d = a + c b + d \dfrac ab + \dfrac cd = \dfrac{a+c}{b+d} is possible for positive integers and derive a contradiction.

First we simplify the LHS by adding the factions (the correct way!) to get

a d + b c b d = a + c b + d \dfrac{ad+bc}{bd}=\dfrac{a+c}{b+d}

Now 'cross multiply'

( a d + b c ) ( b + d ) = a b d + c b d (ad+bc)(b+d)=abd + cbd

Now multiply out the LHS, and some very obvious 'cancelling' gives

a d 2 + c b 2 = 0 ad^2 + cb^2 = 0

Since this is impossible in positive integers, we can invoke the principle of proof by contradiction and say that the bold expression is false, so

a b + c d = a + c b + d \dfrac ab + \dfrac cd = \dfrac{a+c}{b+d} is NOT possible for positive integers.

I really wish I read the questions before answering. Positive integers only...

Kevin Higby - 2 years, 8 months ago

What about all of them =1?

Andres Izquierdo - 2 years, 8 months ago

Log in to reply

Then 1 1 + 1 1 = 2 \frac{1}{1}+\frac{1}{1}=2 while 1 + 1 1 + 1 = 1 \frac{1+1}{1+1}=1 ... Assuming that they would be equal would lead to 1 × 1 2 + 1 × 1 2 = 0 1\times1^2+1\times1^2=0 , or 2=0. Obviously false.

C . - 2 years, 8 months ago

What about all = 0? Then yeah your wrong

German Greyman28 - 2 years, 8 months ago

Log in to reply

Except 0/0 is not a proper fraction... And the problem asked about POSITIVE integers.

C . - 2 years, 8 months ago

Log in to reply

Good point, I admit I looked over that.

German Greyman28 - 2 years, 7 months ago

What about all 1? Your original question suggests a could equal b etc

Paul Marsden - 2 years, 7 months ago

Log in to reply

If they are all 1 the proposed statement reduces to

1+1=1

which is false, isn't it?

Peter Macgregor - 2 years, 7 months ago

Didn't Andres Izquierdo ask this exact same question more than 6 days ago ?! -.-

C . - 2 years, 7 months ago
Andrei Li
Oct 9, 2018

Suppose that there are positive integers a , b , c . d a, b, c. d such that a b + c d = a + c b + d \frac{a}{b}+\frac{c}{d}=\frac{a+c}{b+d} . We know that a b + c d = a d + b c b d \frac{a}{b}+\frac{c}{d}=\frac{ad+bc}{bd} . Then, by equivalence of fractions, a d + b c = a + c ad+bc=a+c b d = b + d bd=b+d The second equation only allows for one solution: b = d = 2 b=d=2 . ( Proof: Suppose that there are two positive integers b , d b,d such that b + d = b d b+d=bd . Subtract d d from both sides and divide by b 1 b-1 to obtain b b 1 = d \frac{b}{b-1}=d ; simplifying the left hand side yields 1 + 1 b 1 = d 1+\frac{1}{b-1}=d . Now, 1 + 1 b 1 Z 1 b 1 Z ( b 1 ) 1 1+\frac{1}{b-1}\in \mathbb{Z}\implies\frac{1}{b-1}\in \mathbb{Z}\implies(b-1)|1 : this requires b 1 = 1 b = 2 b-1=1\implies b=2 . Substituting this into our original equality gives d + 2 = 2 d d = 2 d+2=2d\implies d=2 . For more solutions, see here ) Substituting these values of b , d b,d into the first equation gives 2 a + 2 c = a + c 2 ( a + c ) = a + c a + c = 0 2a+2c=a+c\implies2(a+c)=a+c\implies a+c=0 However, as a a and c c are both positive, this cannot be true. Thus, the answer is yes : there are no positive integers a , b , c , d a,b,c,d such that a b + c d = a + b c + d \frac{a}{b}+\frac{c}{d}=\frac{a+b}{c+d} .

Phil Greene
Oct 8, 2018

a/b+c/d=(a+c)/(b+d) Reduces to (ad+cb)*(b+d)=abd+cbd Reduces to (b/d)^2=-(a/c). ! Impossible

What if a,b,c,d is taken as 1

Ved Mundhe - 2 years, 8 months ago
Frank Bader
Oct 12, 2018

Please note that 0 is positiv too! So we have one solution: a=c=0.

No, 0 is neither positive nor negative.

Pi Han Goh - 2 years, 8 months ago
Ervyn Manuyag
Oct 8, 2018

Nothing won’t work bcuz that isn’t a rule for fractions

Aditya Ghosh
Oct 14, 2018

Multiply both sides with (b+d) and you get a+ad/b+c+cb/d=a+c or, ad/b+cb/d=0, which is not possible for positive a,b,c,d.

Peter Boyajian
Oct 12, 2018

AFSOC there exist such (nonzero) positive integers. Just take the partial of both sides wrt a or c. WLOG, choose c and then we find after a little algebra that d=b+d, so b is zero, which is a contradiction.

You can't use differentiation like that because there's no function to begin with.

Pi Han Goh - 2 years, 8 months ago
Wieter Jacobs
Oct 10, 2018

Multiply both sides by a + c, and the solution is immediately clear

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...