Find the largest real root of
1 0 0 0 x 3 − 1 2 5 4 x 2 − 4 9 6 x + 1 9 1
Details and Assumptions
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.
@mehul is most of all your problem set are you smarter than me? newtons raphoson seems the key to solving the problems
Let's make the substitution x = 3 0 0 0 y + 1 2 5 4 to delete the quadratic term:
1 0 0 0 ( 3 0 0 0 y + 1 2 5 4 ) 3 − 1 2 5 4 ( 3 0 0 0 y + 1 2 5 4 ) 2 − 4 9 6 ( 3 0 0 0 y + 1 2 5 4 ) + 1 9 1 = 0
Expand, clear denominators and simplify:
y 3 − 9 1 8 1 5 4 8 y − 4 3 8 4 7 2 6 1 2 8 = 0
Now, try to match that equation with the identity ( u + v ) 3 − 3 u v ( u + v ) − ( u 3 + v 3 ) = 0 , an equation system will be formed:
y = u + v 3 u v = 9 1 8 1 5 4 8 u 3 + v 3 = 4 3 8 4 7 2 6 1 2 8 . . . ( 1 ) . . . ( 2 ) . . . ( 3 )
Divide ( 2 ) by 3 and cube both sides:
u 3 v 3 = 2 8 6 6 7 1 1 3 2 9 7 1 6 7 4 6 8 0 9 6
Now, make an equation in z with roots u 3 and v 3 :
( z − u 3 ) ( z − v 3 ) = 0 ⟹ z 2 − ( u 3 + v 3 ) z + u 3 v 3 = 0
z 2 − 4 3 8 4 7 2 6 1 2 8 z + 2 8 6 6 7 1 1 3 2 9 7 1 6 7 4 6 8 0 9 6 = 0
Using the quadratic formula we get two values of z :
z = 2 1 9 2 3 6 3 0 6 4 ± 6 0 0 0 i 6 6 2 7 9 6 0 4 1 4 6 6
Hence:
u = 3 2 1 9 2 3 6 3 0 6 4 + 6 0 0 0 i 6 6 2 7 9 6 0 4 1 4 6 6
v = 3 2 1 9 2 3 6 3 0 6 4 − 6 0 0 0 i 6 6 2 7 9 6 0 4 1 4 6 6
Using complex numbers let's take the principal cube roots for u and v :
u = 3 ∣ u ∣ c i s ( 3 ar g u ) ≈ 1 6 2 2 . 6 9 8 9 5 8 + 6 5 3 . 7 3 0 9 0 1 i
v = 3 ∣ v ∣ c i s ( 3 ar g v ) ≈ 1 6 2 2 . 6 9 8 9 5 8 − 6 5 3 . 7 3 0 9 0 1 i
From the equation system, there are three possible values for ( u , v ) . They are: ( u , v ) , ( u w , v w 2 ) and ( u w 2 , v w ) , where w is any primitive 3rd root of unity. Can you check the other values that don't work?
So, we have three solutions for y :
y 1 = u + v ≈ 3 2 4 5 . 3 9 7 9 1 6
y 2 = u w + v w 2 ≈ − 2 7 5 4 . 9 9 4 0 9 3
y 3 = u w 2 + v w ≈ − 4 9 0 . 4 0 3 8 2 3 1
Hence:
x 1 = 3 0 0 0 y 1 + 1 2 5 4 ≈ 1 . 4 9 9 7
x 2 = 3 0 0 0 y 2 + 1 2 5 4 ≈ − 0 . 5 0 0 3
x 3 = 3 0 0 0 y 3 + 1 2 5 4 ≈ 0 . 2 5 4 5
From these solutions, the largest is 1 . 4 9 9 7 .
How u get the the w and also the root (uw, vw^2) and (uw^2, vw)!?
The best solution
Problem Loading...
Note Loading...
Set Loading...
I used Newton-Raphson.
First I tried to find the regions where the function changes sign and thus has a root.
f ( − 1 ) = − 1 0 0 0 − 1 2 5 4 + 4 9 6 + 1 9 1 < 0
f ( 0 ) = 1 9 1 > 0
f ( 1 ) = 1 0 0 0 − 1 2 5 4 − 4 9 6 + 1 9 1 < 0
f ( 2 ) = 8 0 0 0 − 5 0 1 6 − 9 9 2 + 1 9 1 > 0
so we have one root in the region [ − 1 , 0 ] , one root in the region [ 0 , 1 ] and the last one in region [ 1 , 2 ] . We want this last root.
Newton-Rhapson says that the next approximation for the root is x n + 1 according to the formula:
x n + 1 = x n − f ′ ( x n ) f ( x n ) = x n − 3 0 0 0 x n 2 − 2 5 0 8 x n − 4 9 6 1 0 0 0 x n 3 − 1 2 5 4 x n 2 − 4 9 6 x n + 1 9 1
Starting with x 1 = 1 . 5 , the middle of the interval [ 1 , 2 ] , we get the following very fast converging series:
x 1 = 1 . 5
x 2 = 1 . 5 − f ′ ( 1 . 5 ) f ( 1 . 5 ) = 1 . 5 − 2 4 9 2 0 . 5 ≈ 1 . 4 9 9 7 9 9
x 3 = 1 . 4 9 9 7 9 9 − f ′ ( 1 . 4 9 9 7 9 9 ) f ( 1 . 4 9 9 7 9 9 ) = 1 . 4 9 9 7 9 9 − 2 4 9 0 . 6 9 7 5 5 0 . 0 0 0 1 3 0 6 6 6 9 ≈ 1 . 4 9 9 7 9 9
So the solution is rounded to 4 decimals: 1 . 4 9 9 8 .