find the value of 1/(a+bc)+1/(b+ca)+1/(c+ab).
we have
a+b+c=1 ...1
ab+bc+ca=2 .. 2
abc=3 ...3
so a b c are roots of equation
x^3-x^2 + 2x - 3 = 0
further
as a+ b+ c = 1
so a = 1- (b+c)
so a + bc = 1-(b+c) + bc = (1-b)(1-c)
similarly b+ ca = (1-c)(1-a)
c + ba = (1-b)(1-a)
we need to find 1/(a+bc)+1/(b+ca)+1/(c+ab) = 1/ ((1-b)(1-c)) + 1/ ((1-c)(1-a)) + 1/ ((1-b)(1-a))
= (1-a) + (1-b) + (1-c)/(1-a)(1-b)(1-c) = ((a-1) + (b-1) + (c-1))/(a-1)(b-1)(c-1)
so we need to form an equation whose roots are a-1 , b- 1 and c- 1
as a b and c are roots of f(x) = x^3-x^2 + 2x - 3
so a-1 b- 1 and c- 1 are roots of
f(x+1) = (x+1)^3 - (x+1)^2 + 2(x+1) - 3 = x^3 + 2 x^2 + 3x -1
so (a-1) + (b-1) + (c-1) = - 2
(a-1)(b-1)(c-1) = 1
so given expression = ((a-1) + (b-1) + (c-1))/((a-1)(b-1)(c-1)) = - 2
some short and selected math problems of different levels in random order I try to keep the ans simple
Sunday, November 28, 2010
Tuesday, November 16, 2010
2010/063) prove the inequality a+b/c+b+c/a+c+a/b>6 for a,b,c > 0 .?
this can be proved in at least 2 ways
method 1
we need to prove
(a+b)/c+(b+c)/a+(c+a)/b>6
or (a+b+c)/c + (a+b+c)/a + (a+b+ c)/b > 6
or (a+b+c) ( 1/a + 1/b+ 1/c) > 9
we know by AM GM inequality (a+b + c)/ 3 >= (abc)^(1/3)
again (1/a+ 1/b+ 1/c)/3 > = (1/abc)^(1/3)
multiplying we get (a+b + c)/ 3 * (1/a+ 1/b+ 1/c)/3 > = 1
or (a+b + c) * (1/a+ 1/b+ 1/c) >= 9
it is equal when a = b = c
proved
method 2
x+ 1/x >= 2 as (x+1/x) = (sqrt(x) -1/sqrt(x))^2 + 2
puttiing x = a/b . b/c and c/a in 3 times we get 3 eqautions
a/b + b/a >= 2
a/c+ c/a > = 2
b/c + c/b > = 2
adding above 3 and rearranging the terms we get the result.
method 1
we need to prove
(a+b)/c+(b+c)/a+(c+a)/b>6
or (a+b+c)/c + (a+b+c)/a + (a+b+ c)/b > 6
or (a+b+c) ( 1/a + 1/b+ 1/c) > 9
we know by AM GM inequality (a+b + c)/ 3 >= (abc)^(1/3)
again (1/a+ 1/b+ 1/c)/3 > = (1/abc)^(1/3)
multiplying we get (a+b + c)/ 3 * (1/a+ 1/b+ 1/c)/3 > = 1
or (a+b + c) * (1/a+ 1/b+ 1/c) >= 9
it is equal when a = b = c
proved
method 2
x+ 1/x >= 2 as (x+1/x) = (sqrt(x) -1/sqrt(x))^2 + 2
puttiing x = a/b . b/c and c/a in 3 times we get 3 eqautions
a/b + b/a >= 2
a/c+ c/a > = 2
b/c + c/b > = 2
adding above 3 and rearranging the terms we get the result.
Friday, November 12, 2010
2010/062) What are all the rational numbers x such that x + 1/x is an integer?
x + 1/x = (x^2 + 1)/x
This will be an integer if x^2 + 1 = kx, for some integer k. Equivalently,
x^2 - kx + 1 = 0
==> x = [ k +/- sqrt(k^2 - 4) ] / 2
for this to be an integer sqrt(k^2 - 4) has to be an integer say h( same parity as k that is both odd or both even)
k^2-4 = h^2
k^2 - h^2 = 4
so (k-h)(k+h) = 4 = 1 * 4 and 2* 2 and -1 * -4 and -2 * - 2
so k+ h = k-h = 2 or k-h = k+ h = - 2
k+ h = k-h = 2 => k = 2 giving x = 1
k+ h = k-h = 2 => k = -2 giving x = -1
so x = 1 or - 1
This will be an integer if x^2 + 1 = kx, for some integer k. Equivalently,
x^2 - kx + 1 = 0
==> x = [ k +/- sqrt(k^2 - 4) ] / 2
for this to be an integer sqrt(k^2 - 4) has to be an integer say h( same parity as k that is both odd or both even)
k^2-4 = h^2
k^2 - h^2 = 4
so (k-h)(k+h) = 4 = 1 * 4 and 2* 2 and -1 * -4 and -2 * - 2
so k+ h = k-h = 2 or k-h = k+ h = - 2
k+ h = k-h = 2 => k = 2 giving x = 1
k+ h = k-h = 2 => k = -2 giving x = -1
so x = 1 or - 1
2010/061) prove the equality 2 + tan1° tan2° = cot1° tan2°
We know
Tan 2° = tan (1°+1°) = 2 tan 1°/(1- tan ^2 1°)
So tan 2° – tan 2° tan ^2 1° = 2 tan 1°
dividing by tan 1° we get
So tan 2° / tan 1° – tan 2° tan 1° = 2
Or
tan 2° cot 1° – tan 2° tan 1° = 2
So
tan 2° cot 1° = 2 + tan 2° tan 1°
Tan 2° = tan (1°+1°) = 2 tan 1°/(1- tan ^2 1°)
So tan 2° – tan 2° tan ^2 1° = 2 tan 1°
dividing by tan 1° we get
So tan 2° / tan 1° – tan 2° tan 1° = 2
Or
tan 2° cot 1° – tan 2° tan 1° = 2
So
tan 2° cot 1° = 2 + tan 2° tan 1°
Monday, November 1, 2010
2010/060) Proof of Bezout Identity
Bezout Identity states that
If a and b are integers (not both zero) then there exists integers u and v such that
Gcd(a,b) = au + bv
(Note: u and v are not unique
For example
au + bv = a(u-b) + b(v+a))
this can be used by backtracking the euclid’s equation to find Gcd(a,b) in terms of a b. this can be found in a number of books and is the standard process.
However this can be proved using pigeon hole principle also as below
We know that au is divisible by Gcd(a,b) so au mod b is divisible by Gcd(a,b)
Let gcd(a,b ) = k and
b / gcd(a,b) = t
Now taking an mod b( n from 1 to b/gcd(a,b)-1 that is t -1 ) there are
t-1 remainders
they are kn1, kn2, kn3, so on
there are t-1 remainders and all are divisible by k
no remainder can be zero
because na (n from 1 to t-1) cannot be a product of b.
no 2 remainder can be same if they are then difference is a multiple of b
so there has to a n such that one of the remainder is k
as all t- 1 remainders has to be different and values from 1 to n-1 so one remainder has to be 1
an = k mod b or an + bm = k
(u = n, v= m satisfy the condition)
If a and b are integers (not both zero) then there exists integers u and v such that
Gcd(a,b) = au + bv
(Note: u and v are not unique
For example
au + bv = a(u-b) + b(v+a))
this can be used by backtracking the euclid’s equation to find Gcd(a,b) in terms of a b. this can be found in a number of books and is the standard process.
However this can be proved using pigeon hole principle also as below
We know that au is divisible by Gcd(a,b) so au mod b is divisible by Gcd(a,b)
Let gcd(a,b ) = k and
b / gcd(a,b) = t
Now taking an mod b( n from 1 to b/gcd(a,b)-1 that is t -1 ) there are
t-1 remainders
they are kn1, kn2, kn3, so on
there are t-1 remainders and all are divisible by k
no remainder can be zero
because na (n from 1 to t-1) cannot be a product of b.
no 2 remainder can be same if they are then difference is a multiple of b
so there has to a n such that one of the remainder is k
as all t- 1 remainders has to be different and values from 1 to n-1 so one remainder has to be 1
an = k mod b or an + bm = k
(u = n, v= m satisfy the condition)
Subscribe to:
Posts (Atom)