Scribd Logo
Upload

Welcome to Scribd!

  • 29 views

    The PQR Method 2

    Uploaded by

    rohitnsep20261

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    The PQR Method 2

    Uploaded by

    rohitnsep20261
    29 views4 pages

    Original Title

    The Pqr Method 2

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    29 views4 pages

    The PQR Method 2

    Uploaded by

    rohitnsep20261

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 4

    258/ THE pqr METHOD: PART II

    The pqr Method: Part II


    Steven Chow, Howard Halim and Victor Rong

    Part I of this article appears in Crux 43(5).


    The first part of this article discussed the pqr Lemma. In this part, we will examine
    its extension. For completeness, we re-state the pqr Lemma here:
    The pqr Lemma. For three complex numbers a, b and c, let p = a + b + c,
    q = ab + bc + ca, r = abc, and define

    T (p, q, r) = −4p3 r + p2 q 2 + 18pqr − 4q 3 − 27r2 = (a − b)2 (b − c)2 (c − a)2 .

    When we fix two of p, q, r such that there exist triples (p, q, r) satisfying p, q, r ≥
    0 and T (p, q, r) ≥ 0, the unfixed variable obtains its maximum and minimum
    values when two of a, b, c are equal. There is one exception – when r is the
    unfixed variable, its minimum value occurs when either two of a, b, c are equal, or
    one of them is equal to 0.

    Special Conditions
    Consider the following problem:
    Let a, b, c be non-negative numbers such that a2 + b2 + c2 + abc = 4. Prove that
    »
    a + b + c ≥ 2 + abc(4 − a − b − c).

    Here, the condition a2 + b2 + c2 + abc = 4 is equivalent to p2 − 2q + r = 4. However,


    fixing two of p, q, r would fix the third, so we cannot apply the pqr lemma directly.
    Fortunately, we can extend the pqr method to deal with conditions like these.
    The pqr Lemma For Special Conditions. Let the variables a, b, c obey a
    symmetric condition that can be written as G(p, q, r) = 0, where G is a continuous
    function. Let (x, y, z) be a permutation of (p, q, r). Fix some value z ≥ 0 such that
    there exist triples (p, q, r) satisfying p, q, r ≥ 0, T (p, q, r) ≥ 0, and G(p, q, r) = 0.
    Assume that the condition G is equivalent to y = f (x) (while z is fixed), where a
    set of valid values of x is bounded, and f is continuous over that set. If z = r, x
    obtains its maximum and minimum values when two of a, b, c are equal. If z 6= r,
    x obtains its maximum and minimum values when either two of a, b, c are equal,
    or one of them equals 0.
    We will prove this for (x, y, z) = (q, p, r), because the proofs for the others are
    similar.
    Proof. When r is fixed, each of the inequalities T (f (q), q, r) = T (p, q, r) ≥ 0
    and f (q) = p ≥ 0 defines a union of several intervals and rays. The set of valid

    Crux Mathematicorum, Vol. 43(6), June 2017


    Steven Chow, Howard Halim and Victor Rong /259

    values of q is the intersection of those sets, so it is a union of several intervals (it


    cannot contain rays because we are assuming that this set is bounded). At the
    endpoints of those intervals, either T (p, q, r) = 0 or p = 0. The maximum and
    minimum values of q must be at an endpoint. Therefore q attains its maximum
    and minimum values when two of a, b, c are equal. 2
    Now, we can use this to solve the problem above.
    Solution. The inequality
    »
    a+b+c≥2+ abc(4 − a − b − c)
    p
    is equivalent to p ≥ 2 + r(4 − p), and the condition is equivalent to

    p2 − 4 + r
    q= .
    2
    p
    If we fix r, q is a continuous function of p. Let f (p) = p − 2 − r(4 − p). Then,
    the inequality can be written as f (p) ≥ 0. Since f (p) is monotonic, it suffices
    to prove the inequality for the minimum value of p. From the above lemma, this
    occurs when two of the variables are equal.
    WLOG assume that a = b. The condition becomes

    2a2 + c2 + a2 c = 4, or (c + 2)(c + a2 − 2) = 0,

    so c = 2 − a2 . Substituting this into the inequality, we get a4 (a − 1)2 ≥ 0. Since it


    is true for the minimum value of p, it is true for all values of p, and we are done.
    Below is an additional example to further show the usefulness of the pqr method:
    Example. Let a, b, c be non-negative real numbers such that

    a2 + b2 + c2 = ab + bc + ca + (abc − 1)2 .

    Prove that
    ab + bc + ca + 3 ≥ 2(a + b + c).
    p2 −(r−1)2
    Solution. The condition is equivalent to q = 3 . Plugging in this value of
    q into the desired inequality yields

    p2 − (r − 1)2
    + 3 − 2p ≥ 0.
    3
    When p is fixed, this is a concave function in terms of r, and q is a continuous
    function of r. Therefore, it is only necessary to consider the extreme values of r.
    By the pqr lemma for special conditions, r takes an extreme value when WLOG
    a = 0 or a = b.
    If a = 0, then the condition is b2 + c2 = bc + 1, which is equivalent to

    (b + c)2 = 3bc + 1.

    Copyright c Canadian Mathematical Society, 2017


    260/ THE pqr METHOD: PART II

    By the AM-GM inequality,

    3(b + c)2
    (b + c)2 = 3bc + 1 ≤ + 1.
    4
    From this inequality, b + c ≤ 2. The desired inequality is

    bc + 3 ≥ 2(b + c).
    (b+c)2 −1
    By the condition, bc = 3 , so it is enough to prove that

    (b + c)2 − 1
    + 3 ≥ 2(b + c).
    3
    This inequality is equivalent to (2 − b − c)(4 − b − c) ≥ 0, which is true since
    b + c ≤ 2.
    If a = b, then the desired inequality is equivalent to a2 + 2ac + 3 ≥ 4a + 2c. The
    condition becomes
    (a − c)2 = (a2 c − 1)2 ,
    which means that a − c = a2 c − 1 or a − c = 1 − a2 c. If a − c = a2 c − 1, then
    a+1
    c= .
    a2 + 1
    Plugging this into the inequality yields
    Å ã Å ã
    a+1 a+1
    a2 + 2a 2 + 3 ≥ 4a + 2 2 .
    a +1 a +1

    This is equivalent to (a − 1)4 ≥ 0, which is clearly true.


    If a − c = 1 − a2 c, then this is equivalent to (a − 1)(ac + c + 1) = 0. Since a and
    c are non-negative, ac + c + 1 cannot be 0, so a must equal 1 and c can be any
    non-negative real number. Plugging this into the inequality yields 2c + 4 ≥ 2c + 4,
    which always holds true.
    Since we have proved the inequality for when a = 0 and a = b, we are done.

    Problems with Special Conditions


    The following problems may be solved using the pqr lemma for special conditions.
    Problem 5. Let a, b, c be non-negative real numbers such that

    (a + b)(b + c)(c + a) = 8.

    Prove that
    (a + b + c)3 + 5abc ≥ 32.

    Crux Mathematicorum, Vol. 43(6), June 2017


    Steven Chow, Howard Halim and Victor Rong /261

    Problem 6. Let a, b, c be non-negative real numbers such that


    Å ã
    1 1 1
    (a + b + c) + + = 10.
    a b c

    Prove that
    9 a2 + b2 + c2 6
    ≤ ≤ .
    8 ab + bc + ca 5

    Problem 7. Let a, b, c be positive real numbers such that

    2(a + b + c) = a2 b + ab2 + b2 c + bc2 + c2 a + ca2 .

    Prove that
    a2 b2 c2
    + + ≥ 1.
    2 + bc 2 + ca 2 + ab

    Acknowledgements
    We would like to thank our instructors at the Summer Conference: A. Doledenok,
    M. Fadin, A. Menshchikov, and A. Semchankau, for their excellent support during
    the conference. Also, huge thanks to Professor Kumar Murty and the Department
    of Mathematics at the University of Toronto for the generous financial support
    which allowed us to participate in this unique international learning experience.
    Last but not least, many thanks to Olga Zaitseva and Professor Victor Ivrii for
    running the Tournament of Towns program in Toronto, making all the travel
    arrangements, and overseeing us in Russia, otherwise none of this would have
    been possible.

    Copyright c Canadian Mathematical Society, 2017

    You might also like