User:IssaRice/Taking inf and sup separately: Difference between revisions

From Machinelearning
No edit summary
Line 44: Line 44:


Then we have <math>\overline{\int}_I f = \inf(A)</math> and <math>\underline{\int}_I f = \sup(B)</math>. To apply the trick all we need to do is to let <math>g</math> be a p.c. function on <math>I</math> that majorizes <math>f</math>, and let <math>h</math> be a p.c. function on <math>I</math> that minorizes <math>f</math>, and show that <math>p.c.\int_I g\geq p.c.\int_I h</math>.
Then we have <math>\overline{\int}_I f = \inf(A)</math> and <math>\underline{\int}_I f = \sup(B)</math>. To apply the trick all we need to do is to let <math>g</math> be a p.c. function on <math>I</math> that majorizes <math>f</math>, and let <math>h</math> be a p.c. function on <math>I</math> that minorizes <math>f</math>, and show that <math>p.c.\int_I g\geq p.c.\int_I h</math>.
==References==
After I wrote this page, I found the same theorem in Apostol's ''Calculus'' (volume 1, 2nd edition, p. 28) in the section "Fundamental properties of the supremum and infimum".

Revision as of 07:07, 21 June 2019

This page describes a trick that is sometimes helpful in analysis.

Satement

Let and be bounded subsets of the real line. Suppose that for every and we have . Then .

Actually, do and have to be bounded? I think they can even be empty!

Proof

Let and be arbitrary. We have by hypothesis . Since is arbitrary, we have that is an upper bound of the set , so taking the superemum over we have (remember, is the least upper bound, whereas is just another upper bound). Since was arbitrary, we see that is a lower bound of the set . Taking the infimum over , we have , as required.

Applications

liminf vs limsup

(Notation from Tao's Analysis I.)

Let be a sequence of real numbers. Let and let . Then we have .

Consider the sequences and defined by and .

Now consider the sets and . If we can show that for arbitrary , then we can apply the trick to these sets to conclude that .

Lower and upper Riemann integral

(Notation from Tao's Analysis I.)

Let be a bounded interval on the real line, and let .

We have

We want to show .

Define

Then we have and . To apply the trick all we need to do is to let be a p.c. function on that majorizes , and let be a p.c. function on that minorizes , and show that .

References

After I wrote this page, I found the same theorem in Apostol's Calculus (volume 1, 2nd edition, p. 28) in the section "Fundamental properties of the supremum and infimum".