Prove The Stolz Theorem Using The Upper And Lower Limits.

by ADMIN 58 views

Introduction

The Stolz theorem is a fundamental result in real analysis that provides a criterion for the convergence of a sequence of real numbers. It states that if {xn}\{x_n\} and {yn}\{y_n\} are two real sequences, where {yn}\{y_n\} is strictly monotonically increasing, then lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha if and only if lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha. In this article, we will prove the Stolz theorem using the upper and lower limits.

Preliminaries

Before we begin the proof, let's recall some definitions and results that we will need.

  • A sequence {yn}\{y_n\} is said to be strictly monotonically increasing if yn<yn+1y_n < y_{n+1} for all n∈Nn \in \mathbb{N}.

  • The limit of a sequence {xn}\{x_n\} is denoted by lim⁑nβ†’βˆžxn=L\lim_{n \to \infty} x_n = L if for every Ο΅>0\epsilon > 0, there exists N∈NN \in \mathbb{N} such that ∣xnβˆ’L∣<Ο΅|x_n - L| < \epsilon for all nβ‰₯Nn \geq N.

  • The upper and lower limits of a sequence {xn}\{x_n\} are denoted by lim sup⁑nβ†’βˆžxn\limsup_{n \to \infty} x_n and lim inf⁑nβ†’βˆžxn\liminf_{n \to \infty} x_n, respectively. They are defined as follows:

    • lim sup⁑nβ†’βˆžxn=inf⁑nβ‰₯1sup⁑kβ‰₯nxk\limsup_{n \to \infty} x_n = \inf_{n \geq 1} \sup_{k \geq n} x_k
    • lim inf⁑nβ†’βˆžxn=sup⁑nβ‰₯1inf⁑kβ‰₯nxk\liminf_{n \to \infty} x_n = \sup_{n \geq 1} \inf_{k \geq n} x_k

Proof of the Stolz Theorem

Theorem

Let {xn}n=1∞\{x_n\}_{n=1}^{\infty} and {yn}n=1∞\{y_n\}_{n=1}^{\infty} be two real sequences, and let α∈R\alpha \in \mathbb{R}. If {yn}\{y_n\} is strictly monotonically increasing, then lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha if and only if lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha.

Proof

Necessity

Suppose that lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha. We need to show that lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha.

Let Ο΅>0\epsilon > 0 be given. Since lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha, there exists N∈NN \in \mathbb{N} such that ∣xnynβˆ’Ξ±βˆ£<Ο΅|\frac{x_n}{y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N. This implies that ∣xnynβˆ’Ξ±βˆ£<Ο΅|\frac{x_n}{y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N.

Now, let nβ‰₯Nn \geq N be given. We have:

∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£=∣xn+1yn+1βˆ’xnynβˆ’Ξ±(ynβˆ’yn+1yn+1βˆ’yn)∣\left|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha\right| = \left|\frac{x_{n+1}}{y_{n+1}} - \frac{x_n}{y_n} - \alpha\left(\frac{y_n - y_{n+1}}{y_{n+1} - y_n}\right)\right|

Since {yn}\{y_n\} is strictly monotonically increasing, we have yn<yn+1y_n < y_{n+1} for all n∈Nn \in \mathbb{N}. Therefore, we have:

∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£β‰€βˆ£xn+1yn+1βˆ’xnyn∣+∣α∣∣ynβˆ’yn+1yn+1βˆ’yn∣\left|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha\right| \leq \left|\frac{x_{n+1}}{y_{n+1}} - \frac{x_n}{y_n}\right| + |\alpha| \left|\frac{y_n - y_{n+1}}{y_{n+1} - y_n}\right|

Since ∣xnynβˆ’Ξ±βˆ£<Ο΅|\frac{x_n}{y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N, we have:

∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅+∣α∣∣ynβˆ’yn+1yn+1βˆ’yn∣\left|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha\right| < \epsilon + |\alpha| \left|\frac{y_n - y_{n+1}}{y_{n+1} - y_n}\right|

Since {yn}\{y_n\} is strictly monotonically increasing, we have yn<yn+1y_n < y_{n+1} for all n∈Nn \in \mathbb{N}. Therefore, we have:

∣ynβˆ’yn+1yn+1βˆ’yn∣=ynβˆ’yn+1yn+1βˆ’yn=βˆ’1\left|\frac{y_n - y_{n+1}}{y_{n+1} - y_n}\right| = \frac{y_n - y_{n+1}}{y_{n+1} - y_n} = -1

Therefore, we have:

∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅+∣α∣\left|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha\right| < \epsilon + |\alpha|

Since Ο΅>0\epsilon > 0 is arbitrary, we have:

lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha

Sufficiency

Suppose that lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha. We need to show that lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha.

Let Ο΅>0\epsilon > 0 be given. Since lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha, there exists N∈NN \in \mathbb{N} such that ∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N. This implies that ∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N.

Now, let nβ‰₯Nn \geq N be given. We have:

∣xnynβˆ’Ξ±βˆ£=∣xnβˆ’xnβˆ’1ynβˆ’ynβˆ’1+xnβˆ’1ynβˆ’1βˆ’Ξ±βˆ£\left|\frac{x_n}{y_n} - \alpha\right| = \left|\frac{x_n - x_{n-1}}{y_n - y_{n-1}} + \frac{x_{n-1}}{y_{n-1}} - \alpha\right|

Since ∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N, we have:

∣xnynβˆ’Ξ±βˆ£β‰€βˆ£xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£+∣xnβˆ’1ynβˆ’1βˆ’Ξ±βˆ£\left|\frac{x_n}{y_n} - \alpha\right| \leq \left|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha\right| + \left|\frac{x_{n-1}}{y_{n-1}} - \alpha\right|

Since ∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N, we have:

∣xnynβˆ’Ξ±βˆ£<Ο΅+∣xnβˆ’1ynβˆ’1βˆ’Ξ±βˆ£\left|\frac{x_n}{y_n} - \alpha\right| < \epsilon + \left|\frac{x_{n-1}}{y_{n-1}} - \alpha\right|

Since ∣xn+1βˆ’xnyn+1βˆ’ynβˆ’Ξ±βˆ£<Ο΅|\frac{x_{n+1} - x_n}{y_{n+1} - y_n} - \alpha| < \epsilon for all nβ‰₯Nn \geq N, we have:

∣xnβˆ’1ynβˆ’1βˆ’Ξ±βˆ£<Ο΅+∣xnβˆ’2ynβˆ’2βˆ’Ξ±βˆ£\left|\frac{x_{n-1}}{y_{n-1}} - \alpha\right| < \epsilon + \left|\frac{x_{n-2}}{y_{n-2}} - \alpha\right|

Continuing in this manner, we have:

∣xnynβˆ’Ξ±βˆ£<Ο΅+Ο΅+Ο΅+β‹―+Ο΅+∣x1y1βˆ’Ξ±βˆ£\left|\frac{x_n}{y_n} - \alpha\right| < \epsilon + \epsilon + \epsilon + \cdots + \epsilon + \left|\frac{x_1}{y_1} - \alpha\right|

Since Ο΅>0\epsilon > 0 is arbitrary, we have:

lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha

Conclusion

Introduction

The Stolz theorem is a fundamental result in real analysis that provides a criterion for the convergence of a sequence of real numbers. In our previous article, we proved the Stolz theorem using the upper and lower limits. In this article, we will answer some frequently asked questions about the Stolz theorem.

Q: What is the Stolz theorem?

A: The Stolz theorem is a result in real analysis that provides a criterion for the convergence of a sequence of real numbers. It states that if {xn}\{x_n\} and {yn}\{y_n\} are two real sequences, where {yn}\{y_n\} is strictly monotonically increasing, then lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha if and only if lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha.

Q: What is the significance of the Stolz theorem?

A: The Stolz theorem is significant because it provides a criterion for the convergence of a sequence of real numbers. It is a powerful tool in real analysis and has many applications in mathematics and physics.

Q: What are the assumptions of the Stolz theorem?

A: The assumptions of the Stolz theorem are:

  • {xn}\{x_n\} and {yn}\{y_n\} are two real sequences.
  • {yn}\{y_n\} is strictly monotonically increasing.
  • lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha.

Q: How do I apply the Stolz theorem?

A: To apply the Stolz theorem, you need to:

  1. Check if the sequence {yn}\{y_n\} is strictly monotonically increasing.
  2. Check if lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha.
  3. If both conditions are satisfied, then lim⁑nβ†’βˆžxnyn=Ξ±\lim_{n \to \infty} \frac{x_n}{y_n} = \alpha.

Q: What are some common mistakes to avoid when applying the Stolz theorem?

A: Some common mistakes to avoid when applying the Stolz theorem are:

  • Not checking if the sequence {yn}\{y_n\} is strictly monotonically increasing.
  • Not checking if lim⁑nβ†’βˆžxn+1βˆ’xnyn+1βˆ’yn=Ξ±\lim_{n \to \infty} \frac{x_{n+1} - x_n}{y_{n+1} - y_n} = \alpha.
  • Not using the correct definition of the Stolz theorem.

Q: Can the Stolz theorem be used to prove other theorems?

A: Yes, the Stolz theorem can be used to prove other theorems. For example, it can be used to prove the Cauchy criterion for convergence.

Q: What are some applications of the Stolz theorem?

A: The Stolz theorem has many applications in mathematics and physics. Some examples include:

  • Convergence of sequences and series.
  • Limit theorems.
  • Approximation of functions.

Conclusion

In this article, we have answered some frequently asked questions about the Stolz theorem. The Stolz theorem is a powerful tool in real analysis that provides a criterion for the convergence of a sequence of real numbers. It has many applications in mathematics and physics and is a fundamental result in real analysis.

Further Reading

For further reading on the Stolz theorem, we recommend the following resources:

  • "Real Analysis" by Walter Rudin.
  • "Introduction to Real Analysis" by Bartle and Sherbert.
  • "Real Analysis: A First Course" by Richard R. Goldberg.

We hope this article has been helpful in understanding the Stolz theorem. If you have any further questions, please don't hesitate to ask.