Uniform Convergence Examples

Similar documents
Uniform Convergence Examples

Differentiating Series of Functions

Linear Systems of ODE: Nullclines, Eigenvector lines and trajectories

Uniform Convergence and Series of Functions

Linear Systems of ODE: Nullclines, Eigenvector lines and trajectories

Integration and Differentiation Limit Interchange Theorems

Integration and Differentiation Limit Interchange Theorems

Fourier Sin and Cos Series and Least Squares Convergence

Upper and Lower Bounds

Getting Started With The Predator - Prey Model: Nullclines

Dirchlet s Function and Limit and Continuity Arguments

Solving systems of ODEs with Matlab

Consequences of Continuity

Hölder s and Minkowski s Inequality

Riemann Integration. Outline. James K. Peterson. February 2, Riemann Sums. Riemann Sums In MatLab. Graphing Riemann Sums

Lower semicontinuous and Convex Functions

Project One: C Bump functions

The SIR Disease Model Trajectories and MatLab

Consequences of Continuity

Riemann Integration. James K. Peterson. February 2, Department of Biological Sciences and Department of Mathematical Sciences Clemson University

Dirchlet s Function and Limit and Continuity Arguments

Predator - Prey Model Trajectories are periodic

Derivatives in 2D. Outline. James K. Peterson. November 9, Derivatives in 2D! Chain Rule

Taylor Polynomials. James K. Peterson. Department of Biological Sciences and Department of Mathematical Sciences Clemson University

Riemann Sums. Outline. James K. Peterson. September 15, Riemann Sums. Riemann Sums In MatLab

Predator - Prey Model Trajectories are periodic

Matrices and Vectors

More On Exponential Functions, Inverse Functions and Derivative Consequences

Cable Convergence. James K. Peterson. May 7, Department of Biological Sciences and Department of Mathematical Sciences Clemson University

Mathematical Induction Again

Mathematical Induction Again

Runge - Kutta Methods for first and second order models

Defining Exponential Functions and Exponential Derivatives and Integrals

Predator - Prey Model Trajectories and the nonlinear conservation law

Convergence of Sequences

6.2 Deeper Properties of Continuous Functions

Bolzano Weierstrass Theorems I

Matrix Solutions to Linear Systems of ODEs

The First Derivative and Second Derivative Test

The First Derivative and Second Derivative Test

Fourier Sin and Cos Series and Least Squares Convergence

5.5 Deeper Properties of Continuous Functions

Convergence of Sequences

Lecture Notes 3 Convergence (Chapter 5)

The Existence of the Riemann Integral

MATH 140B - HW 5 SOLUTIONS

Distributive property and its connection to areas

A Simple Protein Synthesis Model

Constrained Optimization in Two Variables

General Power Series

Project Two. Outline. James K. Peterson. March 27, Cooling Models. Estimating the Cooling Rate k. Typical Cooling Project Matlab Session

Project Two. James K. Peterson. March 26, Department of Biological Sciences and Department of Mathematical Sciences Clemson University

Hölder s and Minkowski s Inequality

More Protein Synthesis and a Model for Protein Transcription Error Rates

Convergence of Fourier Series

Riemann Integration Theory

Constrained Optimization in Two Variables

The Limit Inferior and Limit Superior of a Sequence

Runge - Kutta Methods for first order models

Proofs Not Based On POMI

Math 104: Homework 7 solutions

Geometric Series and the Ratio and Root Test

Runge - Kutta Methods for first order models

MA 1125 Lecture 15 - The Standard Normal Distribution. Friday, October 6, Objectives: Introduce the standard normal distribution and table.

Solving Linear Systems of ODEs with Matlab

Newton s Cooling Model in Matlab and the Cooling Project!

Advanced Protein Models

Complex Numbers. James K. Peterson. September 19, Department of Biological Sciences and Department of Mathematical Sciences Clemson University

Complex Numbers. Outline. James K. Peterson. September 19, Complex Numbers. Complex Number Calculations. Complex Functions

Lecture 5b: Starting Matlab

Advanced Protein Models

The Method of Undetermined Coefficients.

Proofs Not Based On POMI

Problem Set 1: Solutions Math 201A: Fall Problem 1. Let (X, d) be a metric space. (a) Prove the reverse triangle inequality: for every x, y, z X

Convergence Concepts of Random Variables and Functions

MATH 124B: HOMEWORK 2

Functional Analysis F3/F4/NVP (2005) Homework assignment 3

HOMEWORK 1 SOLUTIONS

More Least Squares Convergence and ODEs

Geometric Series and the Ratio and Root Test

Outline. Additional Nonlinear Systems. Abstract. Finding Equilibrium Points Numerically. Newton s Method

ECE 301 Fall 2011 Division 1. Homework 1 Solutions.

Integration by Parts Logarithms and More Riemann Sums!

Exam 2 Study Guide: MATH 2080: Summer I 2016

5.5 Deeper Properties of Continuous Functions

Extreme Values and Positive/ Negative Definite Matrix Conditions

Constructing Potential Energy Diagrams

More Series Convergence

Exam 2 Solutions October 12, 2006

Lecture 1 From Continuous-Time to Discrete-Time

Probability and Measure

Derivatives and the Product Rule

Function Space and Convergence Types

Lesson 12: Position of an Accelerating Object as a Function of Time

Matrix Completion from Fewer Entries

Section 3.4 Normal Distribution MDM4U Jensen

1.2 Functions and Their Properties Name:

Regression and Covariance

Qualitative Analysis of Tumor-Immune ODE System

Math 651 Introduction to Numerical Analysis I Fall SOLUTIONS: Homework Set 1

Transcription:

Uniform Convergence Examples James K. Peterson Department of Biological Sciences and Department of Mathematical Sciences Clemson University October 13, 2017 Outline More Uniform Convergence Examples

Example Let (xn) be the sequence of functions on [0, 1] defined by xn(t) = 2nt. e nt2 Discuss pointwise and uniform convergence on [0, 1]. First, it is easy to see the pointwise limit function is x(t) = 0 on [0, 1]. Is the convergence uniform? Let s begin by graphing some of these functions in Octave. The code is straightforward. 1 T = l i n s p a c e ( 0, 1, 5 1 ) ; f = @( t, n ) 2 n t. / exp ( n t. ˆ 2 ) ; p l o t (T, f (T, 5 ),T, f (T, 1 0 ),T, f (T, 2 0 ),T, f (T, 3 0 ) ) ; x l a b e l ( t ) ; y l a b e l ( y ) ; t i t l e ( x_n ( t ) = 2 nt / e ^{ nt ^2} for n = 5,10,20 and 30 ) ; 6 l e g e n d ( x5, x10, x20, x30, location, north ) ; Here you see a plot of x 5, x 10, x 20 and x 30 on the interval [0, 1]. You can clearly see the peaks of the functions are increasing with the maximums occuring closer and closer to 0.

The derivative here, after some simplification, is x n(t) = 2n(1 2nt2 ) is zero at t = 0 and t = ±1/ 2n. The critical point at t = 0 is uninteresting and the maximum occurs at tn = 1/ 2n and has value 2n/e. Let s calculate xn x here. We have xn x = sup 2nt/e nt2 0 = 2n/e 0 t 1 e nt2 which For the convergence to be uniform, given ɛ > 0, we would have to be able to find Nɛ so that xn x < ɛ when n > Nɛ. Here that means we want 2n/e < ɛ when n > Nɛ. But for n > Nɛ, 2n/e. So this cannot be satisfied and the convergence is not uniform. Now let s look at this same sequence on a new interval. Example Examine the convergence of the sequence xn(t) = 2nt on [ 2, 2]. nt2 We can graph some of the functions in this sequence on this new interval using the code below. T = l i n s p a c e ( 2,2,101) ; f = @( t, n ) 2 n t. / exp ( n t. ˆ 2 ) ; p l o t (T, f (T, 5 ),T, f (T, 1 0 ),T, f (T, 2 0 ),T, f (T, 3 0 ),T, f (T, 5 0 ) ) ; 4 x l a b e l ( t ) ; y l a b e l ( y ) ; t i t l e ( x_n ( t ) = 2 nt / e ^{ nt ^2} on [ -2,2] ) ; e

Here you see a plot of some of these functions on the interval [ 2, 2]. You can clearly see the peaks of the functions are increasing with the minimums and maximums occuring closer and closer to 0. The minimum occur at 1/ 2n with value 2n/e, while the maximums occur at 1/ 2n with value 2n/e. Now take a small positive number a and mentally imagine drawing a vertical line through previous picture at that point. The maximum s occur at 1/ 2n. If 1/ 2n <.1 or n > 50, the maximum values all occur before the value x =.1. We generate a plot for this as follows: (here we do not add the code for the axis labels etc.) T = l i n s p a c e ( 0. 0 5, 0. 5, 1 0 1 ) ; f = @( t, n ) 2 n t. / exp ( n t. ˆ 2 ) ; p l o t (T, f (T, 5 0 0 ),T, f (T, 7 0 0 ),T, f (T, 9 0 0 ) ) ; On the interval [0.1, 0.5], the functions have their maximum value at xn(0.1) =.2n/e.01n. Since.2n/e.01n 0 as n, we see given ɛ > 0, there is Nɛ so that.2n/e.01n < ɛ when n > Nɛ. You can see this behavior clearly in the next figure.

There we graph x500, x700 and x900 and you can easily see the value of these functions at 0.1 is decreasing. Hence, for n > Nɛ, xn 0 < ɛ unif and so we can say xn 0 on [.1, r] for any r > 0.1. A similar analysis works for any a > 0. unif In fact, if we looked at the other side, we would show xn 0 on any interval of the form [ r, a] with a > 0. Indeed, this convergence is uniform on any interval [c, d] as long as 0 [c, d].

Now let s look at this same sequence on [.1, 2] analytically. We know the minimum of xn(t) occur at 1/ 2n with value 2n/e, while the maximum of xn(t) occurs at 1/ 2n with value 2n/e. There is an N so that n > N implies < 1/ 2n <.1; i.e. the maximum value before x =.1. It occurs before the interval [.1, 2]. Since on the right of the maximum, xn(t) decreases, this tells us the maximum of xn(t) on [.1, 2] is given by xn(.1). So sup t [.1.2] xn(t) 0 = xn(.1) =.2n e.01n and we see this goes to zero with n. So convergence is uniform on [.1, 2]. Example Examine the pointwise and uniform convergence of (xn) where xn(t) = 3nt/e 4nt2 on intervals of R. Work this out in class following the example just done. Use MatLab and blackboard sketches to see what is going on.

Example Examine the pointwise and uniform convergence of (xn) where xn(t) = 2nt/e 3nt2 on intervals of R. Work this out in class following the example just done. Use MatLab and blackboard sketches to see what is going on. Homework 21 21.1 Let xn be defined by xn(t) = { 0, 0 t 1/n n, 1/n < t < 2/n 0, 2/n t 1 Determine if (xn) converges uniformly to its limit function. 21.2 Examine the pointwise and uniform convergence of (xn) where xn(t) = 6nt/e 2nt2 on intervals of R. Do a careful analysis just like we have done for the other examples. Sketches are required!