4 3 4 + 1 − 4 3 4 − 1 1
Find the nearest integer to the given expression above.
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.
Great method!
Will you kindly explain the double conjugate?
Log in to reply
It is all about rationalization, I hope you know it.
It is the process of elimination the radical forms from the equation by multiplying with conjugates and use the perfect square differences of x 2 − y 2 = ( x + y ) ( x − y ) . Here, I did it twice in order to eliminate it completely.
I used a linear approximation of the roots. This is likely to work because 1 < < 3 4 .
First, divide numerator and denominator by 3: 4 1 + ( 3 1 ) 4 − 4 1 + ( 3 1 ) 4 1 / 3 . Next, apply the approximation k 1 + x ≈ 1 + k x + O ( x 2 ) ; this results in ( 1 + 4 1 ⋅ ( 3 1 ) 4 ) − ( 1 − 4 1 ⋅ ( 3 1 ) 4 ) 1 / 3 = 2 ⋅ 4 1 ⋅ ( 3 1 ) 4 1 / 3 = 2 ⋅ 3 3 = 5 4 . How good is this approximation? Note that the remaining terms in the approximation are of the order O ( x 2 ) , that is of the order ( 8 1 1 ) 2 ∼ 1 0 − 4 . Thus we expect our solution to be accurate well within 1%. This justifies concluding that 54 is indeed the closest integer.
Direct calculation bears this out: a more exact value obtained by calculator is 5 3 . 9 9 8 1 9 9 5 1 .
Nice use of the appropriate approximation to try and bound it.
However, note that "order O ( x 2 ) " doesn't mean that the error is bounded by ∣ x 2 ∣ . Instead, what we have is that the error is bounded by a ∣ x 2 ∣ for some a . It could be a very large value of a , as opposed to a = 1 .
You will have to use the remainder term of the Taylor series, to know what the value of a is. In this case, it is close to 2 k 2 ( k − 1 ) .
Log in to reply
Of course. I did not want to make my solution too long. The fact that ( k − 1 ) / 2 k 2 is much smaller than 1, and that subsequent coefficients will be exponentially smaller, shows that we are quite safe here.
A more rigorous method would prove that 1 + k x > k 1 + x > 1 + k x − 2 k 2 ( k − 1 ) x 2 , here 1 + 1 8 1 > 6 1 + 3 1 > 1 + 1 8 1 − 6 4 8 5 .
Problem Loading...
Note Loading...
Set Loading...
When applying the double conjugate this expression will be 2 1 ( 8 2 + 8 0 ) ( 4 8 2 + 4 8 0 ) As 8 1 = 9 ; 4 8 1 = 3 , the closest approximation of the value will be substituted to the actual value, which will produce 2 1 ( 8 2 + 8 0 ) ( 4 8 2 + 4 8 0 ) ≈ 2 1 ( 9 + 9 ) ( 3 + 3 ) ≈ 5 4