34 CHAPTER 1. FUNDAMENTAL CONCEPTS

Let 0 < δ < min

1, |xn−a|2

(1+∑

n−1k=0

(n

k

)|x|k)−1

 Then if |y− x| < δ , from the

above,

(xn −a)(yn −a)≥ |xn −a|2 − |xn −a|2

2> 0

and so xn −a and yn −a have the same sign.

Theorem 1.10.2 Let a > 0 and let n > 1. Then there exists a unique x > 0 such thatxn = a.

Proof: Let S denote those numbers y ≥ 0 such that tn − a < 0 for all t ∈ [0,y]. Nownote that from the binomial theorem,

(1+a)n −a =n

∑k=0

(n

k

)ak1n−k −a ≥ 1+a−a = 1 > 0

Thus S is bounded above by 1+ a and 0 ∈ S. Let x ≡ sup(S). Then by definition of sup,for every δ > 0, there exists t ∈ S with |x− t|< δ .

If xn −a > 0, then by the above lemma, for t ∈ S sufficiently close to x,

(tn −a)(xn −a)> 0

which is a contradiction because the first factor is negative and the second is positive.Hence xn −a ≤ 0. If xn −a < 0, then from the above lemma, there is a δ > 0 such that ift ∈ (x−δ ,x+δ ) ,xn−a and tn−a have the same sign. This is also a contradiction becausethen x ̸= sup(S). It follows xn = a.

From now on, we will use this fact that nth roots exist and are unique whenever it isconvenient to do so.

1.11 Completing the SquareThere is a very important process called completing the square. The idea is as follows. Fora,b,c real numbers with a ̸= 0, it is desired to write the expression ax2 +bx+c in the forma(x− γ)2 + β . I will now show how to do this. It is very important because if you havedone it, you can see that letting x = γ yields the smallest possible value of the expressionax2 +bx+ c for all x a real number provided a > 0 and it yields the largest possible valueif a < 0. Here are the steps for doing it:

1. ax2 +bx+ c = a(x2 + b

a x+ ca

)2. a

(x2 + b

a x+ ca

)= a

(x2 + b

a x+ b2

4a2 − b2

4a2 +ca

)3. a

(x2 + b

a x+ b2

4a2 − b2

4a2 +ca

)= a

((x+ b

2a

)2 −(

b2

4a2 − 4ac4a2

))4. a

((x+ b

2a

)2 −(

b2

4a2 − 4ac4a2

))= a

(x+ b

2a

)2+(

4ac−b2

4a

)The following fundamental theorem gives a formula for finding solutions to a quadratic

equation, one of the form ax2 +bx+ c = 0.

34 CHAPTER 1. FUNDAMENTAL CONCEPTS-1Let 0 < 6 < min | 1, a Gee ( j nt) Then if |y—x| < 6, from theabove,| n ?—_ —__ >(x" a) (y"—a) > |x" —a)?— 50and so x” —a and y” —a have the same sign. JJTheorem 1.10.2 Let a> and let n> 1. Then there exists a unique x > 0 such thatx” =a.Proof: Let S denote those numbers y > 0 such that ¢” — a < 0 for all t € [0,y]. Nownote that from the binomial theorem,n(I+a)"-a=)° ( : JA az ttanant>0k=0Thus S is bounded above by 1+ a and0€S. Let x = sup(S). Then by definition of sup,for every 6 > 0, there exists ¢ € S with |x—t| <6.If x” —a > 0, then by the above lemma, for t € S sufficiently close to x,("—a)(x"—a) >0which is a contradiction because the first factor is negative and the second is positive.Hence x" —a < 0. If x” —a <0, then from the above lemma, there is a 6 > 0 such that ift € (x—6,x+6),x"—aand t” —a have the same sign. This is also a contradiction becausethen x ~ sup(S). It followsx” =a. JFrom now on, we will use this fact that n"” roots exist and are unique whenever it isconvenient to do so.1.11 Completing the SquareThere is a very important process called completing the square. The idea is as follows. Fora,b,c real numbers with a ¥ 0, it is desired to write the expression ax* + bx-+c in the forma(x—7)* + B. I will now show how to do this. It is very important because if you havedone it, you can see that letting x = y yields the smallest possible value of the expressionax* + bx +c for all x a real number provided a > 0 and it yields the largest possible valueif a < 0. Here are the steps for doing it:lL. ax? +bxtco=a(x?+2x4+S)2. a(P+Px42)=a(P+er+- B42)2,b be b _ b\2 RP 43. a(x + 2x4 — 246) =a((x+$) -(4-4))b\2 4 _ b\2 4ac—b?s a((o 8) (Be—$8) <a 8) + (G2)The following fundamental theorem gives a formula for finding solutions to a quadraticequation, one of the form ax? +bx+c=0.