% times noreferences tall pstricks multido fancybox twoup nopagenumbers \begin{center} \bf Some review questions: inner product spaces, eigenvalues \end{center} \vspace{0.2in} \problemno{1} Let $V$ be an inner product space. \par\noindent{\bf(a):\ } State precisely what it means for the inner product to be {\it linear in the first variable}. \par\noindent{\bf(b):\ } State precisely what it means for the inner product to be {\it linear in the second variable}. \par\noindent{\bf (c):\ } Let $u,v,w \in V$. We know that $$\inn{u,v} = 2,\ \inn{v,w} = -3, \inn{u,w} = 5,$$ $$\norm{u} = 1,\ \norm{v} = 2,\ \norm{w} = 7.$$% Compute $\inn{u+v,v+w}$. \vspace{0.1in} \problemno{2} Let $A$ be an $n \times n$ matrix with eigenvalues $\vec \lambda1n$. What statement about the eigenvalues is equivalent to saying that $A$ is invertible? \vspace{0.1in} \problemno{3} Find the coordinates of $\BRMAT{1\cr-1\cr2}$ \WRT\ the basis $$\BRMAT{-3/5\cr4/5\cr0},\ \BRMAT{4/5\cr3/5\cr0},\ \BRMAT{0\cr0\cr1}$$% for $\R^3$. [{\sc Hint:}\ What is special about this basis?] \vspace{0.1in} \problemno{4} Equip $\R^3$ with the inner product given by $$\inn{a,b}\ =\ a_1b_1 + 2a_2b_2 + 4a_3b_3.$$% Let $p = \BRMAT{1\cr2\cr3}$, $q = \BRMAT{3\cr2\cr1}$. With respect to the given inner product, what is the distance between $p$ and $q$? \vspace{0.1in} \problemno{5} Let $V$ be an inner product space with basis $v_1, v_2, v_3, v_4$. You are in the middle of applying the Gram-Schmidt process to find an orthonormal basis $w_1, w_2, w_3, w_4$ for $V$: you have already found $w_1, w_2, w_3$. \vspace{0.1in} \par\noindent{\bf(a):\ } If $x,u \in V$, and $\norm{u} = 1$, give a formula for the {\it component\/} of $x$ in the direction $u$. \par\noindent{\bf(b):\ } Explain {\it precisely\/} how you compute $w_4$. \vspace{0.1in} \problemno{6} Let $A = \BRMAT{1&a\cr0&1}$, where $a$ is nonzero. Show that $A$ is not diagonalizable. \vspace{0.1in} \problemno{7} Let $V$ be the vector space of all continuous functions from $[0,1]$ to $\R$, with the inner product $$\inn{f,g}\ =\ \int_0^1 f(t)g(t)dt.$$% Let $f$ be the element of $V$ given by $f(t) = t^2$. Compute $\norm{f}$. \vspace{0.1in} \problemno{8} Let $A = \BRMAT{1&0\cr6&-1}$. \par\noindent{\bf(a):\ } Find a matrix $P$ which diagonalizes $A$. \par\noindent{\bf(b):\ } What is $P^{-1}AP$? \vspace{0.1in} \problemno{10} Let $A = \BRMAT{3&0\cr8&-1}$. \vspace{0.05in} \par\noindent{\bf(a):}\ find the characteristic equation of $A$ \par\noindent{\bf(b):}\ find the eigenvalues of $A$ \par\noindent{\bf(c):}\ find bases for the eigenspaces of $A$ \vspace{0.1in} \problemno{11} Let $A = \BRMAT{0&-1\cr1&0}$. Find an invertible matrix $P$ and a diagonal matrix $D$ such that $P^{-1}AP = D$. [{\sc Note:}\ There will be complex numbers in your answer.] \vspace{0.1in} \problemno{12} Given square matrices $P$, $A$ of the same size, with $P$ invertible, simplify $(P^{-1}AP)^{10}$. \vspace{0.1in} \problemno{13} Let $V$ be an inner product space. We are given an orthonormal basis $\setof{a,b,c}$ for $V$, and another element $x$ of $V$, which is a unit vector. We know that $\inn{a+b,x} = 5$, $\inn{c,x} = 4$, and $\norm{a+x} = 4$. \par\noindent{\bf(a):}\ Compute $\inn{c,b+c+x}$. \par\noindent{\bf(b):}\ Compute $\inn{a,x}$. \par\noindent{\bf(c):}\ Express $x$ as a linear combination of $a,b,c$. \vspace{0.1in} \problemno{14} Let $V$ be the vector space of all continuous functions fom $[0,1]$ to $\R$. Define an inner product on $V$ by $$\inn{f,g} = \int_0^1 f(t)g(t)dt.$$ Let $h$ be the element of $V$ given by $h(t) = t^3$. \vspace{0.1in} \par\noindent{\bf(a):}\ Compute $\norm{h}$. \par\noindent{\bf(b):}\ Exhibit a basis for the subspace $W$ of $V$ consisting of functions having the form $a+bt$, where $a,b \in \R$. \par\noindent{\bf(c):}\ Exhibit an orthonormal basis for $W$. \vspace{0.1in} \problemno{15} {\bf(a):}\ Exhibit three different $2 \times 2$ orthogonal matrices, all having determinant $1$. \par\noindent {\bf(b):}\ Exhibit two different $3 \times 3$ orthogonal matrices, both having determinant $1$. \vspace{0.1in} \problemno{16} Let $\R^3$ have the standard inner product (dot product). Let $W$ be the subspace of $\R^3$ given by $$W = \left\{ \BRMAT{t\cr t\cr t}: t \in \R \right\}.$$% \par\noindent{\bf(a):\ } Exhibit a nonzero element of $W$. \par\noindent{\bf(b):\ } Find a basis for $W^\perp$. [{\sc Hint.}\ It has dimension $2$.] \vspace{0.1in} \problemno{17} {\bf(a):}\ Let $A,B$ be orthogonal matrices (of the same size). Show that $AB$ is orthogonal. \par\noindent{\bf(b):}\ Show that a $3 \times 3$ real matrix cannot have eigenvalues $i, i, 1$. \vspace{0.1in} \problemno{18} {\bf(a):\ } Let $A$ be an $m \times n$ matrix, and let $b \in \R^m$. State precisely what it means for $x \in \R^n$ to be a {\it least squares solution\/} of the equation $Ax = b$. (Your answer should not involve matrix transpose -- that has to do with how you calculate least squares solutions. Do not be vague. A single sentence will suffice.) \par\noindent{\bf(b):}\ Explain how to find all least squares solutions to $Ax = b$. You may assume that the reader knows how to solve systems of linear equations. Use a complete sentence or sentences. \par\noindent{\bf(c):}\ Under what circumstances is it the case that the solution set of $Ax = b$\\ equals \setofh{least squares solutions to $Ax = b$}? [Perhaps your answer to (a) will help you do this part.] \par\noindent{\bf(d):\ } Find the least squares solution(s) to $Ax = b$, where $$A = \BRMAT{1&1\cr-1&1\cr-1&2},\ \ \ b = \BRMAT{7\cr0\cr-7}.$$ \vspace{0.1in} \problemno{19}\ \ Let $\R^3$ have the Euclidean inner product. Let $W$ be the subspace of $\R^3$ given by $$W\ =\ \left\{ \BRMAT{t\cr-t\cr t}: t \in \R \right\}.$$% \par\noindent{\bf(a):\ } Find a basis for $W$. \par\noindent{\bf(b):\ } Find a basis for $W^\perp$. [{\sc Note:}\ $\dim W^\perp = 2$, so if you find two linearly independent elements of $W^\perp$ (perhaps by inspection), then they will form a basis for $W^\perp$.] \par\noindent{\bf(c):\ } Find an orthonormal basis for $W$. \par\noindent{\bf(d):\ } Find an orthonormal basis for $W^\perp$. \par\noindent{\bf(e):\ } Compute the projection of $\BRMAT{1\cr2\cr3}$ onto $W$. \par\noindent{\bf(f):\ } Compute the projection of $\BRMAT{1\cr2\cr3}$ onto $W^\perp$. You do not have to simplify your answer. \par\noindent [{\sc Hint.}\ \ You can deduce (f) by means of (d) or (e), but it's easiest to use (e).] \vspace{0.1in} \problemno{20}\ \ {\bf(a):\ } Complete the following definition. A square matrix $A$ is \vspace{0.15in} \par\noindent{\it diagonalizable\/} if there exists \blank3 \vspace{0.15in} \par\noindent such that \blank{4.5}. \vspace{0.1in} \par\noindent{\bf(b):\ } Assume that $A$ is diagonalizable. Prove that $A^2$ is diagonalizable. \vspace{0.1in} \problemno{21}\ \ Let $V = \shC[0,1] = \setofh{continuous functions from $[0,1]$ to $\R$}$, with the inner product given by $\inn{f,g} = \int_0^1 f(t)g(t)dt$. Find a positive real number $c$ such that $\norm{ct} = 1$. \vspace{0.1in} \problemno{22}\ \ Let $A$ be a $3 \times 3$ matrix with characteristic polynomial $\lambda^3 - \lambda^2$. \vspace{0.1in} \par\noindent{\bf(a):\ } What are the eigenvalues of $A$? \par\noindent{\bf(b):\ } For each eigenvalue $\lambda$ of $A$, list all possible values for the dimension of the corresponding eigenspace $E_\lambda$. (You do not have to exhibit specific matrices $A$.) \par\noindent{\bf(c):\ } What can you deduce from (b) about the possible values for the nullity of $A$? Explain briefly. \par\noindent{\bf(d):\ } Exhibit a $3 \times 3$ diagonal matrix $A$ with characteristic polynomial $\lambda^3 - \lambda^2$. \vspace{0.1in} \problemno{23} Let $\R^3$ have the standard inner product. Let $$v_1 = \BRMAT{-1\cr0\cr0},\ v_2 = \BRMAT{0\cr3/5\cr4/5}, \ v_3 = \BRMAT{0\cr4/5\cr-3/5},\ w = \BRMAT{1\cr2\cr3}.$$% Then \setof{v_1,v_2,v_3}\ is an orthonormal basis for $\R^3$. Find the coordinates of $w$ \WRT\ this basis. \vspace{0.1in} \problemno{24} Let $P_1$ be the vector space of polynomials of degree $\leq 1$, with the inner product $$\inn{p,q}\ =\ \int_0^1 p(x)q(x)dx.$$% Find an orthonormal basis for $P_1$. \vspace{0.1in} \problemno{25} Let $V = \R^2$ with the standard inner product. Use pictures to describe all orthonormal bases for $V$.