北航矩阵论学习笔记
北京航空航天大学矩阵理论学习笔记，总结版，学霸总结，可以放心下载使用北京航空航天大学张京蕊工程系统工程系月录§0补充公式§1 Jordan(约当)标准形(简介)§2线性变换与矩阵.24§3欧式空间与QR分解.48§4常用矩阵分解●鲁D●●·,,,,,74§5范数与级数.81§6广义逆A..97§7直积拉直及应用105矩阵理论A笔记北京航空航天大学张京蕊工程系统工程系§0补充公式令A=(a)mxn∈C",风x)=4o+a1x0定义f(4)=a0+a1A+…+amAm,其中I=l若g(x)=bo+b1x+…+bkx,(x)g(x)=g(x)(x),则f(4)“g(A)=g(A)f(A)分块公式A10令A,A1,A2为方阵00 A(2)f(A),fx)为多项式令A=,A1,4为方阵AO(2)f(4)相似关系:A∽B,(PAP=B)则:(1)(P1AP)=P!AP,(k=0,1,2,(2)f(PAP)=PfA)P,f(x)为多项式许尔公式( schur):每个复方阼,A-(a)nxm都相似丁上三角形。共113页矩阵理论A笔记第1页北京航空航天大学张京蕊工程系统工程系即:P-1AP=其中41,,的次序可以任意指定Pf:用归纳法n=1时成立可以设为(n=1阶方阵成立对于n阶方阵A=(an)2×n设特征值为A,…,n取为对应的特征向量,记为a1≠0,A1=1ax1把a1扩展为可逆方阵Q=(a1,02,xn)22e又:g(a,a,…,.)=(Qa,Qba2,,Qan)其中Qe1,aQ0Q4=QA(a1a2,…an)2-I(Aa,,AAQ=(Qa,、+)…,(*)其中A1为(n-1所阶0人:0 A为由假设,对于A1必有(n-1)阶P,可推出PAPEg知n阶方阵A,适合A=0,则A+|=1共113页矩阵理论A笔记第2页北京航空航天大学张京蕊工程系统工程系Pf:A=0→任意特征值A=0→>=0即全体特征值为00,,00由需要P1AP=→PAP+7=1pAP+PP|=P(4+1)P=14+1→A+1=-1注(1)若AB(相似),则AB有相同特征值A,可引入记号:谱集(4)={2,2,…,λ}(全体特征值,含重复)A∽B→o()=o(B)(2)A∽B→1-A=1-B-(2-4元一2)…(-n),特征多项式PAP=B=A-A=p(1-A)P=A-B引理:若A0A2,则M-A|-|M1-4|-1-A1|2-A2→ar(4)=o(A)∪a(42k+1,Ak-2,…n1f(x2)设B,f(x)为多项式,则f(B)=o f(,)引理:若n阶方阵A的谱集(4)=1,42,…},则)的全体特社值为)2,…,),x)为多项式Pf:由许尔定理,A∽B→f(4)∽f(B)f(x)的全体特征值为(A1)(42),,()},fx)为多项式例如:4为A的特征值→x为4的特征值。(x)=x)共113页矩阵理论A笔记第3页北京航空航天大学张京蕊工程系统工程系引理:令B,f(x)=x-B|=(x-41)(x-12)….(x-n)则fB)=(B-1D(B-21)…(B-A1D=0Pf:当n=2时,B=0x2f(x)=(x-1)(x-2)000→f(B)-(B-41)(B-21)(2-元)0(00∴得证★ Cayley公式:设n阶方阵A的特征多项式为f(x)=|x-A|=a+a1x+…,+x则f4)=anl+a14+…,+4=0Pf:由许尔PAP=B=→P(4)P=fp3P)=f(B)=0(引理)定义若多项式x)使(4)=0,则称(x)为A的个零化式结论方阵A的特征多项式)=1x1-4为A的一个零化式g特征多项式fx)=x2可知:f(A)=A2+1=+I=00-1Hx)=|xI-A|=(x-)(x+i,(i=√-1,t2=-1)f(A)=(A-i)(4+i1=0也可取P=则PPAP=,对角形共113页矩阵理论A笔记第4页北京航空航天大学张京蕊工程系统工程系g:知A则A"=0Onxn由 Cayley特征多项式:f(x)=x"→f(4)=4"=0Ex 1. A=求P使得PP为对角阵,并验证 Cayley定理2.A=cd/,求fx)=x1-4验证f4)-0补充知识( schur公式、 Cayley公式)应用由A"=-(a0I+a1A+1A·AanA+a142+…+a.,A把①代入②→Am1=(-)+(+)4+…+(+)41可知:任何和(m≥n)都可写成,4,,A的线性组合任何多项式g(A),可写成lA,…,4的组合。Fg:若A|≠0,fx)=xI-A|=a0+a1x+…+x",ao=|-A|≠则A可用A的多项式表示∵a1A+a242+…+an21A-+A"--a072A(a1+a24+…+an-142+A)Aa1+…+an1A"2+A-1零化式定义:若g(x)=b+b1x+…+bnx,使得g(4)=bn+b14+…+bn4m=0,称g(x)为方阵A的零化式注:方阵A的零化式有无穷多个∴取特征多项式x)则4)=0任取式M(x),f(A(4)=0→f(x)(x)也是零化式极小式定义:在方阵A的零化式集合中,去次数最小的且首项系数为1的零化式m(x),称它为A的极小式共113页矩阵理论A笔记第5页北京航空航天大学张京蕊工程系统工程系注:极小式唯一性质:①极小式m(x)必为特征多项式fx)=|xI-A的因式。②特征多项式fx)=|x1-A的每个单因子(x-4)也是极小式的因子)f(x)=|x1-4=(x-x)(x-2)则极小式m(x)=(x-x)(x-2)y…(x-,),且1≤l1≤m1,1≤l2≤m2,…,1≤l≤n,41,A2…,n互不相同210EgA=020,B=020,求极小式mA(),m()解:(1)|xI-A|=(x-2)(x-1)极小式为:(x-2)(x-1)或(x-2)(x-1)计算:(4-2/)4-1)=000010k≠000000∴极小式为m4(x)=(x-2)2(x-1)(2)|-B|-=(x-2)2(x-1)00000计算:(B-2)B-1)=000010=000-1八000∴极小式为m(x)=(x-2)(x-1)Eg求下列极小式m(x)4604-60(1)A=-3-50,(2)B=2-303-6100210(3)C,(4)D=000010002000解:(1)特征多项式|x7-A|-(x-1)(x+2)极小式为:(x-1)(x+2)或(x-1)(x+2)共113页矩阵理论A笔记第6页北京航空航天大学张京蕊工程系统工程系验证:(4-D(A+2D=0∴极小式为m(x)-(x-1)(x+2)(3)解法如下引理:A1,A2的极小式为m1(x),m2(x)A10的极小式m(x)等丁m1(x),m2(x)的最小公倍式0A2(此引力可推广到A1,42,43)0100极小式为(x-1)2,0010极小式为(x-1)0取最小公倍式(x-1)2为C的极小式。460(5)F-/40,A1=020|,A00 A0123-6101O引理;设D=,则D的极小式m(x)O验证:先证D的性质(右推公式)设A-(an)xn=(a1,2,…,n)则有AD=(0,01,a2,,.m1)AD2=(0,0,∞1,,x12)AD=(0,….0.,a1,,axn)单位向量技巧:∵AI=A(en,e2…,en)=(el,leAen)=A=(a1, a2,. a,)∴Ae1=01,Ae2=(2,.,A→AD=A(0,e1,e2,…,en-1)=(0,a1,a2…,an-)同理AD2=(AD)D=(0,.01,.12)可知:D-1-(D)Dy2-(0.,0,,e1)≠0D"=(D)D1=0,而特征多项式(x)=|x1-D|=x,极小式为某个x共113页矩阵理论A笔记第7页
- 2020-12-09下载
- 积分：1
Introduction.to.Stochastic.Processes.with.R
An introduction to stochastic processes through the use of RIntroduction to Stochastic Processes with R is an accessible and well-balanced presentation of the theory of stochastic processes, with an emphasis on real-world applications of probability theory in the natural and social sciences. The uINTRODUCTIONTO STOCHASTICPROCESSES WITH RINTRODUCTIONTO STOCHASTICPROCESSES WITH RROBERT P DOBROWWILEYCopyright o 2016 by John Wiley Sons, Inc. All rights reservedPublished by John Wiley Sons, Inc, Hoboken, New JerseyPublished simultaneously in CanadaNo part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form orby any means, electronic, mechanical, photocopying, recording, scanning, or otherwise, except aspermitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the priorwritten permission of the Publisher, or authorization through payment of the appropriate per-copy fee tothe Copyright Clearance Center, Inc, 222 Rosewood Drive, Danvers, MA,(978)750-8400, fax978)750-4470,oronthewebatwww.copyright.comRequeststothePublisherforpermissionshouldbe addressed to the Permissions Department, John Wiley sons, Inc, lll River Street, Hoboken, NJ07030,(201)748-6011,fax(201)748-6008,oronlineathttp://www.wiley.com/go/permissionsLimit of liability/ Disclaimer of warranty While the publisher and author have used their best efforts inpreparing this book, they make no representations or warranties with respect to the accuracy orcompleteness of the contents of this book and specifically disclaim any implied warranties ofmerchantability or fitness for a particular purpose. No warranty may be created or extended by salesrepresentatives or written sales materials. The advice and strategies contained herein may not be suitablefor your situation. You should consult with a professional where appropriate. Neither the publisher norauthor shall be liable for any loss of profit or any other commercial damages, including but not limited tospecial, incidental, consequential, or other damagesFor general information on our other products and services or for technical support, please contact ourCustomer Care Department within the United States at(800)762-2974, outside the United States at(317)572-3993 or fax(317)572-4002Wiley also publishes its books in a variety of electronic formats. Some content that appears in print maynot be available in electronic formats. For more information about Wiley products, visit our web site atwww.wiley.comLibrary of Congress Cataloging-in-Publication Data:Dobrow. Robert p. authorIntroduction to stochastic processes with r/ Robert P. Dobrowpages cmIncludes bibliographical references and indexISBN978-1-118-74065-1( cloth)1. Stochastic processes. 2. R( Computer program language)I. TitleQC20.7.S8D6320165192′302855133-dc232015032706Set in 10/12pt, Times-Roman by SPi Global, Chennai, IndiaPrinted in the united states of america1098765432112016To my familyCONTENTSPrefaceAcknowledgmentsList of Symbols and Notationabout the companion Website1 Introduction and review1.1 Deterministic and stochastic models. 11. 2 What is a Stochastic Process? 61. 3 Monte Carlo Simulation. 91.4 Conditional Probability, 101. 5 Conditional Expectation, 18Exercises. 342 Markov Chains: First Steps402.1 Introduction. 402.2 Markov Chain Cornucopia, 422.3 Basic Computations, 522. 4 Long-Term behavior-the Numerical evidence, 592.5 Simulation. 652.6 Mathematical Induction*. 68Exercises. 70CONTENTS3 Markov Chains for the long term763.1 Limiting Distrib763.2 Stationary Distribution, 803.3 Can you find the way to state a? 943.4 Irreducible markov Chains. 1033.5 Periodicity, 1063.6 Ergodic Markov Chains, 1093.7 Time Reversibility, 1143.8 Absorbing Chains, 1199 Regeneration and the strong markov property 1333.10 Proofs of limit Theorems*, 135Exercises. 1444 Branching processes1584.1 Introduction. 1584.2 Mean Generation Size. 1604.3 Probability Generating Functions, 1644.4 Extinction is Forever. 168Exercises. 1755 Markov Chain Monte Carlo1815.1 Introduction. 1815.2 Metropolis-Hastings Algorithm, 1875.3 Gibbs Sampler, 1975.4 Perfect Sampling*, 20.55.5 Rate of Convergence: the Eigenvalue Connection*, 2105.6 Card Shuffing and Total Variation Distance. 212Exercises. 2196 Poisson process2236.1 Introduction. 2236.2 Arrival. Interarrival Times. 2276.3 Infinitesimal Probabilities. 2346.4 Thinning, Superposition, 2386.5 Uniform Distribution. 2436.6 Spatial Poisson Process, 2496.7 Nonhomogeneous Poisson Process. 2536.8 Parting Paradox, 255Exercises. 2587 Continuous- Time markov Chains2657.1 Introduction. 265
- 2020-12-10下载
- 积分：1
