Lecture 12 - Monotonic Sequences

See last time for our review of the ALT. See the link for more information.

As a result we get some basic facts:

Facts

  1. If (an)=(c) for some c∈R then liman=c.
  2. If (an)=(a1,a2,...) and suppose liman=a∈R. Then any N∈N has it such that for all n≥N then (an)→a still. See 22 Sequences Practice#2.2.7.

This allows us to prove the Limits and Order (Order Limit Theorem):

Order Limit Theorem

Suppose liman=a and limbn=b. Then:
i. If an≥0 for all n∈N, then a≥0.
ii. If an≤bn for all n∈N then a≤b.
iii. If ∃c∈R where c≤bn for all n∈N, then c≤b. Similarly, if an≤c for all n∈N then a≤c.

Proof

  1. Suppose an≥0 for all n∈N. Assume that a<0 for the contrary. Since an→a then for any n∈N. Then:
    |an−a|<ε=−a⇒a<an−a<−a
    But then looking at the right inequality:
    an−a<−a⇒an<0
    contradicts that an≥0. As a result, a≮0 so then a≥0.

  2. Notice that bn−an≥0 for all n∈N, so then use (1) along with the Limit Laws (Algebraic Limit Theorem) to get that b−a≥0⇒b≥a.

  3. Create the sequence (cn)=c for all n∈N. Then apply (ii) for both cases, giving that since (cn)→c via our Limit Laws (Algebraic Limit Theorem), then c≤b and similarly an≤c.
    ☐

Note

We can make the theorem stronger by only imposing that eventually that these sequences have an≥0 after some N point, for all n≥N. This would be proved in a similar way.
If a property has this idea, then we say that the sequence eventually has this property.

We then covered Monotonic Sequences:

Monotonic Sequences

A Sequence (an) is

  • increasing if ∀n∈N we have an≤an+1
  • decreasing if ∀n∈N we have an≥an+1
  • monotone if it is either increasing or decreasing.

See also Monotone Convergence Theorem.