Q
stringlengths
4
3.96k
A
stringlengths
1
3k
Result
stringclasses
4 values
1. If two functions \( f, g : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) are both integrable, then \( f + g \) is also integrable, and\n\n\[ \n{\int }_{{\mathbb{R}}^{n}}\left( {f + g}\right) \left| {{d}^{n}\mathbf{x}}\right| = {\int }_{{\mathbb{R}}^{n}}f\left| {{d}^{n}\mathbf{x}}\right| + {\int }_{{\mathbb{R}}^{n}}g\le...
1. For any subset \( A \subset {\mathbb{R}}^{n} \), we have\n\n4.1.35\n\nApplying this to each cube \( C \in {\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) \), we get\n\n\[ \n{U}_{N}\left( f\right) + {U}_{N}\left( g\right) \geq {U}_{N}\left( {f + g}\right) \; \geq \;{L}_{N}\left( {f + g}\right) \geq {L}_{N}\left( f\ri...
Yes
1. A bounded function \( f \) with bounded support is integrable if and only if both \( {f}^{ + } \) and \( {f}^{ - } \) are integrable.
1. If \( f \) is integrable, then by part 4 of Proposition 4.1.14, \( \left| f\right| \) is integrable, and so are \( {f}^{ + } = \frac{1}{2}\left( {\left| f\right| + f}\right) \) and \( {f}^{ - } = \frac{1}{2}\left( {\left| f\right| - f}\right) \), by parts 1 and 2. Now assume that \( {f}^{ + } \) and \( {f}^{ - } \) ...
Yes
Proposition 4.1.16. If \( {f}_{1}\left( \mathbf{x}\right) \) is integrable on \( {\mathbb{R}}^{n} \) and \( {f}_{2}\left( \mathbf{y}\right) \) is integrable on \( {\mathbb{R}}^{m} \), then the function\n\n\( g\left( {\mathbf{x},\mathbf{y}}\right) = {f}_{1}\left( \mathbf{x}\right) {f}_{2}\left( \mathbf{y}\right) \; \) o...
Proof. First suppose \( {f}_{1} \) and \( {f}_{2} \) are nonnegative. For any \( {A}_{1} \subset {\mathbb{R}}^{n} \) and \( {A}_{2} \subset {\mathbb{R}}^{m} \), we have\n\n\[{M}_{{A}_{1} \times {A}_{2}}\left( g\right) = {M}_{{A}_{1}}\left( {f}_{1}\right) {M}_{{A}_{2}}\left( {f}_{2}\right) ;\;{m}_{{A}_{1} \times {A}_{2}...
Yes
Lemma 4.1.19. An interval \( I = \left\lbrack {a, b}\right\rbrack \) has length \( \left| {b - a}\right| \) .
Proof. Of the cubes (i.e., intervals) \( C \in {\mathcal{D}}_{N}\left( \mathbb{R}\right) \), at most two contain an endpoint \( a \) or \( b \) . The others are either entirely in \( I \) or entirely outside; on those\n\n\[ \n{M}_{C}\left( {\mathbf{1}}_{I}\right) = {m}_{C}\left( {\mathbf{1}}_{I}\right) = \left\{ \begin...
No
Proposition 4.1.20 (Volume of \( n \) -dimensional parallelogram). The \( n \) -dimensional parallelogram\n\n\[ P\overset{\text{ def }}{ = }{I}_{1} \times \cdots \times {I}_{n} \subset {\mathbb{R}}^{n} \]\n\nformed by the product of intervals \( {I}_{i} = \left\lbrack {{a}_{i},{b}_{i}}\right\rbrack \) has volume\n\n\[ ...
Proof. This follows immediately from Proposition 4.1.16, applied to\n\n\[ {\mathbf{1}}_{P}\left( \mathbf{x}\right) = {\mathbf{1}}_{{I}_{1}}\left( {x}_{1}\right) {\mathbf{1}}_{{I}_{2}}\left( {x}_{2}\right) \ldots {\mathbf{1}}_{{I}_{n}}\left( {x}_{n}\right) \]
Yes
Theorem 4.1.21 (Sum of volumes). If two disjoint sets \( A, B \) in \( {\mathbb{R}}^{n} \) are pavable, then so is their union, and the volume of the union is the sum of the volumes:
Proof. Since \( {\mathbf{1}}_{A \cup B} = {\mathbf{1}}_{A} + {\mathbf{1}}_{B} \) if \( A \) and \( B \) are disjoint, the result follows from part 1 of Proposition 4.1.14.
Yes
Proposition 4.1.22 (Volume invariant under translation). Let \( A \) be any pavable subset of \( {\mathbb{R}}^{n} \) and \( \overrightarrow{\mathbf{v}} \in {\mathbb{R}}^{n} \) any vector. Denote by \( A + \overrightarrow{\mathbf{v}} \) the set \( A \) translated by \( \overrightarrow{\mathbf{v}} \). Then \( A + \overri...
Proof. Let \( {K}_{N} \) be the set of cubes \( C \in {\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) \) with \( C \subset A \) (i.e., the set of cubes at the \( N \) th level that are entirely inside \( A \) ), and let \( {H}_{N} \) be the set of cubes \( C \in {\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) \) such t...
Yes
Proposition 4.1.23 (Set with volume 0). A bounded set \( X \subset {\mathbb{R}}^{n} \) has volume 0 if and only if for every \( \epsilon > 0 \) there exists \( N \) such that
\[ \begin{array}{l} \mathop{\sum }\limits_{{C \in {\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) }}{\operatorname{vol}}_{n}\left( C\right) \leq \epsilon . \\ C \cap X \neq \varnothing \end{array} \]
No
Proposition 4.1.24 (Scaling volume). If \( A \subset {\mathbb{R}}^{n} \) has volume and \( t \in \mathbb{R} \), then \( {tA} \) has volume, and\n\n\[ \n{\operatorname{vol}}_{n}\left( {tA}\right) = {\left| t\right| }^{n}{\operatorname{vol}}_{n}\left( A\right) \n\]
Proof. By Proposition 4.1.20, this is true if \( A \) is a parallelogram, in particular, a cube \( C \in {\mathcal{D}}_{N} \) . Assume \( A \) is a subset of \( {\mathbb{R}}^{n} \) whose volume is well defined. This means that \( {\mathbf{1}}_{A} \) is integrable, or, equivalently, that\n\n\[ \n\mathop{\lim }\limits_{{...
Yes
What is the center of gravity of the right triangle \( T \) shown in Figure 4.2.1?
The area of \( T \), i.e., \( {\int }_{T}{dxdy} \), is \( {ab}/2 \). Using the techniques of Section 4.5, we can easily compute\n\n\[ \n{\int }_{T}{xdxdy} = \frac{{a}^{2}b}{6}\;\text{ and }\;{\int }_{T}{ydxdy} = \frac{a{b}^{2}}{6} \]\n\nso by equation 4.2.1, the center of gravity of \( T \) is the point \( \overline{\m...
Yes
Example 4.2.4 (Computing \( \pi \) using Buffon’s needle). Toss a needle of length 1 on a piece of lined paper, with lines parallel to the \( x \) -axis spaced 1 apart. The sample space should be the space of positions of the needle. We will be interested in the probability that the needle intersects a line, so we will...
So the probability that the needle will intersect a line is \[ \mathbf{P}\left( A\right) = \frac{2}{\pi }{\int }_{A}\left| {d\theta ds}\right| = \frac{2}{\pi }{\int }_{0}^{\pi }\frac{1}{2}\sin {\theta d\theta } = \frac{2}{\pi }. \] Thus the probability of intersecting the line is \( 2/\pi \) . This provides a (cumberso...
Yes
Theorem 4.2.7 (The central limit theorem). If an experiment and a random variable have expectation \( E \) and standard deviation \( \sigma \), then if the experiment is repeated \( n \) times, with average result \( \bar{x} \), the probability that \( \overline{x}\; \) is between \( \;E + \frac{\sigma }{\sqrt{n}}a\; \...
\[ \frac{1}{\sqrt{2\pi }}{\int }_{a}^{b}{e}^{-{y}^{2}/2}{dy} \]
Yes
What is the probability that a fair coin tossed 1000 times will come up heads between 510 and 520 times?
In principle, this is straightforward: just compute the sum\n\n\[ \frac{1}{{2}^{1000}}\mathop{\sum }\limits_{{k = {510}}}^{{520}}\left( \begin{matrix} {1000} \\ k \end{matrix}\right) \]\n\nIn practice, computing these numbers would be extremely cumbersome; it is much easier to use the central limit theorem. Our individ...
Yes
Theorem 4.3.1 (Criterion for integrability). A function \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) is integrable if and only if it is bounded with bounded support, and for all \( \epsilon > 0 \), there exists \( N \) such that ![489cc797-1a02-4514-b519-0c9454f84c47_441_1.jpg](images/489cc797-1a02-4514-b519-0c945...
In inequality 4.3.1 we sum the volume of only those cubes for which the oscillation of the function is more than epsilon. If, by making the cubes very small (choosing \( N \) sufficiently large) the sum of their volumes is less than epsilon, then the function is integrable: we can make the difference between the upper ...
No
Theorem 4.3.6. Any continuous function \( {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) with bounded support is integrable.
Proof. Theorem 4.3.6 follows almost immediately from Theorem 1.6.11. Let \( f \) be continuous with bounded (hence compact) support. Since \( \sup \left| f\right| \) is realized on the support, \( f \) is bounded. By Theorem 1.6.11, \( f \) is uniformly continuous: choose \( \epsilon \) and find \( \delta > 0 \) such t...
Yes
Corollary 4.3.7. Let \( X \subset {\mathbb{R}}^{n} \) be compact and let \( f : X \rightarrow \mathbb{R} \) be continuous. Then the graph \( {\Gamma }_{f} \subset {\mathbb{R}}^{n + 1} \) has volume 0.
Proof. Since \( X \) is compact, it is bounded, and there is a number \( A \) such that for all \( N \), the number of cubes \( C \in {\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) \) such that \( X \cap C \neq \varnothing \) is at most \( A{2}^{nN} \) . Choose \( \epsilon > 0 \), and use Theorem 1.6.11 to find \( \de...
Yes
Corollary 4.3.8. Let \( U \subset {\mathbb{R}}^{n} \) be open and let \( f : U \rightarrow \mathbb{R} \) be a continuous function. Then any compact part \( Y \) of the graph of \( f \) has \( \left( {n + 1}\right) \) - dimensional volume 0.
Proof. The projection of \( Y \) into \( {\mathbb{R}}^{n} \) is compact, and the restriction of \( f \) to that projection satisfies the hypotheses of Corollary 4.3.7.
No
Theorem 4.3.9. A function \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \), bounded with bounded support, is integrable if it is continuous except on a set of volume 0 .
Proof. Denote by \( \Delta \) (\
No
Corollary 4.3.10. Let \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) be integrable, and let \( g : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) be a bounded function. If \( f = g \) except on a set of volume 0, then \( g \) is integrable, and \[ {\int }_{{\mathbb{R}}^{n}}f\left| {{d}^{n}\mathbf{x}}\right| = {\int }_{{...
Proof. The support of \( g \) is bounded, since the support of \( f \) is bounded and so is the support of \( g - f \) . For any \( N \), a cube \( C \in {\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) \) where \( {\operatorname{osc}}_{C}\left( g\right) > \epsilon \) is either one where \( {\operatorname{osc}}_{C}\left...
Yes
Corollary 4.3.12. Any polynomial function \( p \) can be integrated over any set \( A \) of finite volume; that is, \( p \cdot {\mathbf{1}}_{A} \) is integrable.
Proof. The function \( p \cdot {\mathbf{1}}_{A} \) meets the conditions of Theorem 4.3.9: it is bounded with bounded support and is continuous except on the boundary of \( A \), which has volume 0 .
Yes
Example 4.4.3 (A set with measure 0, undefined volume). The set of rational numbers in the interval \( \left\lbrack {0,1}\right\rbrack \) has measure 0.
You can list them in order, for instance, as \( 1,1/2,1/3,2/3,1/4,2/4,3/4,1/5,\ldots \) (The list is infinite and includes some numbers more than once.) Center an open interval of length \( \epsilon /2 \) at 1, an open interval of length \( \epsilon /4 \) at \( 1/2 \), an open interval of length \( \epsilon /8 \) at \(...
Yes
Theorem 4.4.4 (A countable union of sets of measure 0 has measure 0). Let \( i \mapsto {X}_{i} \) be a sequence of sets of measure 0 . Then\n\n\( {X}_{1} \cup {X}_{2} \cup \ldots \; \) is a set of measure 0 .
Proof of Theorem 4.4.4: We can turn the sequences\n\n\[ \n{B}_{1, i},\ldots ,{B}_{j, i},\ldots \n\]\n\ninto a single sequence by listing the boxes in some order, just as we did for the rational numbers in Example 4.4.3.\n\nFor instance, one could list first the boxes where \( i + j = 2 \), then those where \( i + j = 3...
No
Example 4.4.3 (and, more generally, any countable set) corresponds to the case of Theorem 4.4.4 where each \( {X}_{i} \) is a single point.
Proof of Theorem 4.4.4. Since there exist infinite sequences \( i \mapsto {B}_{j, i} \) of boxes (one sequence for each \( j \) ) such that\n\n\[ \n{X}_{1} \subset {B}_{1,1} \cup {B}_{1,2} \cup \ldots ,\;\text{ and }\;\sum \operatorname{vol}{B}_{1, i} \leq \frac{\epsilon }{2} \n\]\n\n\[ \n{X}_{2} \subset {B}_{2,1} \cup...
Yes
Corollary 4.4.5. Let \( {B}_{R}\left( \mathbf{0}\right) \) be the ball of radius \( R \) centered at \( \mathbf{0} \) . If for all \( R \geq 0 \) the subset \( X \subset {\mathbb{R}}^{n} \) satisfies \( {\operatorname{vol}}_{n}\left( {X \cap {B}_{R}\left( \mathbf{0}\right) }\right) = 0 \), then \( X \) has measure 0.
Proof. Since \( X = { \cup }_{m = 1}^{\infty }\left( {X \cap {B}_{m}\left( \mathbf{0}\right) }\right) \), it is a countable union of sets of volume 0 , hence measure 0 .
Yes
Proposition 4.4.6. Let \( X \) be a subspace of \( {\mathbb{R}}^{n} \) of dimension \( k < n \) . Then \( X \) has measure 0, and any translate of \( X \) has measure 0 .
Proof. The first statement follows from Proposition 4.3.5 and Corollary 4.4.5; the second from Proposition 4.1.22.
No
Consider the function\n\n\[ f\left( x\right) = \left\{ \begin{array}{ll} \frac{1}{q} & \text{ if }x = \frac{p}{q}\text{ is rational, written in lowest terms with }q > 0\text{ and }\left| x\right| \leq 1 \\ 0 & \text{ if }x\text{ is irrational or }\left| x\right| > 1. \end{array}\right. \]\n\nThis function is integrable...
For instance, \( f\left( {3/4}\right) = 1/4 \), but arbitrarily close to \( 3/4 \) we have irrational numbers, giving \( f\left( x\right) = 0 \). But such values form a set of measure 0. The function is continuous at the irrationals: arbitrarily close to any irrational number \( x \) you will find rational numbers \( p...
No
Theorem 4.4.8 (What functions are integrable). Let \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) be bounded with bounded support. Then \( f \) is integrable if and only if it is continuous except on a set of measure 0.
Proof. We will start with the harder direction: if \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \), bounded with bounded support, is continuous almost everywhere, then it is integrable. We will use the criterion for integrability given by Theorem 4.3.1; thus we want to prove that for all \( \epsilon > 0 \) there exis...
Yes
Lemma 4.4.9. Let \( i \mapsto {B}_{i} \) be a sequence satisfying \( \Delta \subset \cup {B}_{i} \) and \( \sum {\operatorname{vol}}_{n}{B}_{i} < \epsilon \) . Then \( {\operatorname{osc}}_{{B}_{i}}\left( f\right) > \epsilon \) on only finitely many boxes \( {B}_{i} \) .
Proof of Lemma 4.4.9. Assume the lemma is false. Then there exist an infinite subsequence of boxes \( j \mapsto {B}_{{i}_{j}} \) and two infinite sequences of points, \( j \mapsto {\mathbf{x}}_{j}, j \mapsto {\mathbf{y}}_{j} \) in \( {B}_{{i}_{j}} \), such that \( \left| {f\left( {\mathbf{x}}_{j}\right) - f\left( {\mat...
Yes
Corollary 4.4.11. Let \( f \) and \( g \) be integrable functions on \( {\mathbb{R}}^{n} \) such that\n\n\( f \geq g \) and \( \int f\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| = \int g\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| \) . Then\n\n\[ \n\{ \mathbf{x} \mid f\left( \mathbf{x}\right...
Proof. The function \( f - g \) is integrable, hence continuous almost everywhere. Thus if \( f > g \) on a set not of measure 0, there exists \( {\mathbf{x}}_{0} \) such that \( f\left( {\mathbf{x}}_{0}\right) > g\left( {\mathbf{x}}_{0}\right) \) and \( f - g \) is continuous at \( {\mathbf{x}}_{0} \) . Then there exi...
Yes
Corollary 4.4.12 (Product of integrable functions is integrable). If two functions \( f, g : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) are both integrable, then \( {fg} \) is also integrable.
Proof. If \( f \) is continuous except on a set \( {X}_{f} \) of measure 0, and \( g \) is continuous except on a set \( {X}_{g} \) of measure 0, then (Proposition 1.5.29) \( {fg} \) is continuous except on a subset of \( {X}_{f} \cup {X}_{g} \) . By Theorem 4.4.4, this subset has measure 0 . Thus by Theorem 4.4.8, \( ...
Yes
Corollary 4.4.13 (Integration is translation invariant). If a function \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) is integrable, then for any \( \overrightarrow{\mathbf{v}} \in {\mathbb{R}}^{n} \), the function \( \mathbf{x} \mapsto f\left( {\mathbf{x} - \overrightarrow{\mathbf{v}}}\right) \) is integrable.
Corollary 4.4.13 follows from Theorem 4.4.8.
Yes
Now let’s integrate an unspecified function \( f : {\mathbb{R}}^{2} \rightarrow \mathbb{R} \) over the area bordered on the top by the parabolas \( y = {x}^{2} \) and \( y = {\left( x - 2\right) }^{2} \) and on the bottom by the straight lines \( y = - x \) and \( y = x - 2 \), as shown in Figure 4.5.4.
Let's start again by sweeping our pencil from left to right, which corresponds to the outer integral being with respect to \( x \) . The limits for the outer integral are clearly \( x = 0 \) and \( x = 2 \), giving\n\n\[ \n{\int }_{0}^{2}\left( {\int {fdy}}\right) {dx} \n\]\n\nAs we sweep our pencil from left to right,...
No
Example 4.5.3 (Setting up a multiple integral in \( {\mathbb{R}}^{3} \) ). Already in \( {\mathbb{R}}^{3} \) this kind of visualization becomes much harder. Suppose we want to integrate a function over the pyramid \( P \) shown in Figure 4.5.5 and given by the formula\n\n\[ P = \left\{ {\left. {\left( \begin{array}{l} ...
There are six ways to apply Fubini's theorem, which in this case because of symmetries will result in the same expressions with the variables permuted. Let us vary \( z \) first. For instance, we can lift a piece of paper and see how it intersects the pyramid at various heights. Clearly the paper will only intersect th...
No
In the preceding examples, we could deduce the upper and lower limits from pictures of the region over which we want to integrate. Here is a case where the picture doesn't give the answer. Let us set up a multiple integral over the ellipse \( E \) defined by \( E = \\left\\{ {\\left. \\left( \\begin{array}{l} x \\\\ y ...
The boundary of \( E \) is defined by \( {y}^{2} + {xy} + {x}^{2} - 1 = 0 \), so on the boundary, \( y = \\frac{-x \\pm \\sqrt{{x}^{2} - 4\\left( {{x}^{2} - 1}\\right) }}{2} = \\frac{-x \\pm \\sqrt{4 - 3{x}^{2}}}{2}. Such values exist only if \( 4 - 3{x}^{2} \\geq 0 \), i.e., \( \\left| x\\right| \\leq 2\\sqrt{3}/3 \)....
Yes
Let \( f : {\mathbb{R}}^{2} \rightarrow \mathbb{R} \) be the function \( f\left( \begin{array}{l} x \\ y \end{array}\right) = {xy}{\mathbf{1}}_{S}\left( \begin{array}{l} x \\ y \end{array}\right) \), where \( S \) is the unit square, as shown in Figure 4.5.7. Then \[ {\int }_{{\mathbb{R}}^{2}}f\left( \begin{array}{l} x...
\[ = {\int }_{0}^{1}{\left\lbrack \frac{{x}^{2}y}{2}\right\rbrack }_{x = 0}^{x = 1}{dy} = {\int }_{0}^{1}\frac{y}{2}{dy} = \frac{1}{4}.\;\bigtriangleup \]
Yes
Let us integrate the function \( {e}^{-{y}^{2}} \) over the triangle shown in Figure 4.5.8:\n\n\[ T = \left\{ {\left. {\left( \begin{array}{l} x \\ y \end{array}\right) \in {\mathbb{R}}^{2}}\right| \;0 \leq x \leq y \leq 1}\right\} .
Fubini's theorem gives us two ways to write this integral as an iterated one-dimensional integral:\n\n\[ {\int }_{0}^{1}\left( {{\int }_{x}^{1}{e}^{-{y}^{2}}{dy}}\right) {dx}\;\text{ and }\;{\int }_{0}^{1}\left( {{\int }_{0}^{y}{e}^{-{y}^{2}}{dx}}\right) {dy}. \]\n\nThe first cannot be computed in elementary terms, sin...
Yes
Example 4.5.7 (Volume of a ball in \( {\mathbb{R}}^{n} \) ). Let \( {B}_{R}^{n}\left( \mathbf{0}\right) \) be the ball of radius \( R \) in \( {\mathbb{R}}^{n} \), centered at \( \mathbf{0} \), and let \( {b}_{n}\left( R\right) \) be its volume. By Proposition 4.1.24, \( {b}_{n}\left( R\right) = {R}^{n}{b}_{n}\left( 1\...
By Fubini's theorem,\n\n\[ \underset{\begin{matrix} \text{vol. of} \\ \text{unit ball in }{\mathbb{R}}^{n} \end{matrix}}{\underbrace{{\beta }_{n}}} = {\int }_{{B}_{1}^{n}\left( \mathbf{0}\right) }\left| {{d}^{n}\mathbf{x}}\right| = {\int }_{-1}^{1}\overset{1}{\overbrace{\left( {\int }_{{B}_{\sqrt{1 - {x}_{n}^{2}}}^{n -...
Yes
Exercise 4.5.10 asks you to show that if two points \( \\mathbf{x},\\mathbf{y} \) are chosen in the unit square, without privileging any part of the square, then \( E\\left( {\\left| \\mathbf{x} - \\mathbf{y}\\right| }^{2}\\right) = 1/3 \)
\[ E\\left( {\\left( x - y\\right) }^{2}\\right) = {\\int }_{0}^{1}{\\int }_{0}^{1}{\\left( x - y\\right) }^{2}{dxdy} = {\\int }_{0}^{1}\\left( {{\\int }_{0}^{1}{x}^{2} - {2xy} + {y}^{2}{dx}}\\right) {dy} \]\n\n\[ = {\\int }_{0}^{1}{\\left\\lbrack \\frac{{x}^{3}}{3} - {x}^{2}y + {y}^{2}x\\right\\rbrack }_{0}^{1}{dy} = ...
No
Theorem 4.5.10 (Fubini’s theorem). Let \( f : {\mathbb{R}}^{n} \times {\mathbb{R}}^{m} \rightarrow \mathbb{R} \) be an integrable function, and suppose that for each \( \mathbf{x} \in {\mathbb{R}}^{n} \), the function \( \mathbf{y} \mapsto f\left( {\mathbf{x},\mathbf{y}}\right) \) is integrable. Then the function
\[ \mathbf{x} \mapsto {\int }_{{\mathbb{R}}^{m}}f\left( {\mathbf{x},\mathbf{y}}\right) \left| {{d}^{m}\mathbf{y}}\right| \] is integrable, and \[ {\int }_{{\mathbb{R}}^{n + m}}f\left( {\mathbf{x},\mathbf{y}}\right) \left| {{d}^{n}\mathbf{x}}\right| \left| {{d}^{m}\mathbf{y}}\right| = {\int }_{{\mathbb{R}}^{n}}\left( {{...
Yes
Proposition 4.6.4 (Product rules). If \( {f}_{1},\ldots ,{f}_{n} \) are functions that are integrated exactly by an integration rule, i.e., \[ {\int }_{a}^{b}{f}_{j}\left( x\right) {dx} = \mathop{\sum }\limits_{i}{w}_{i}{f}_{j}\left( {p}_{i}\right) \;\text{ for }j = 1,\ldots, n \] then the product \[ f\left( \mathbf{x}...
Proof. This follows immediately from Proposition 4.1.16. Indeed, \[ {\int }_{{\left\lbrack a, b\right\rbrack }^{n}}f\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| = \left( {{\int }_{a}^{b}{f}_{1}\left( {x}_{1}\right) d{x}_{1}}\right) \ldots \left( {{\int }_{a}^{b}{f}_{n}\left( {x}_{n}\right) d{x}_{n}}\right...
Yes
Each weight for the two-dimensional Simpson's rule is the product of two one-dimensional Simpson weights from Definition 4.6.1. In the very simple case where we wish to integrate over a square, dividing it into only four subsquares, and sampling the function at each vertex, we have nine samples in all.
Let us do this with the square of sidelength \( b - a = 2 \) . Since \( n = 1 \) (this square corresponds, in the one-dimensional case, to the first piece in Figure 4.6.2), the one-dimensional weights are\n\n\[ \n{w}_{1} = {w}_{3} = \frac{b - a}{6n} \cdot 1 = \frac{1}{3}\;\text{ and }\;{w}_{2} = \frac{b - a}{6n} \cdot ...
Yes
In Example 4.5.9 we computed the expected value for the determinant of a \( 2 \times 2 \) matrix. Now let try the same for \( 3 \times 3 \) matrices, using MATHEMATICA’s Montecarlo program to approximate\n\n\[{\int }_{C}\left| {\det A}\right| \left| {{d}^{9}\mathbf{x}}\right|\]\n\n4.6.25\n\ni.e., to evaluate the averag...
Four runs of length 50000 gave the following estimates of the integral:\n\n\[{.13453},{.13519},{.13409},{.13456}.\]\n\n4.6.26\n\nHow accurate are these guesses? We might expect the first two digits to be correct, and the third to be off by 1 or 2 . What does the theory say?\n\nThe same program calculates the standard d...
Yes
Example 4.8.2 (The function \( {\Delta }_{3} \) ). If\n\n\[ A = \left\lbrack \begin{array}{lll} 1 & 3 & 4 \\ 0 & 1 & 1 \\ 1 & 2 & 0 \end{array}\right\rbrack \text{, then}{A}_{\left\lbrack 2,1\right\rbrack } = \left\lbrack \begin{array}{lll} 1 & 3 & 4 \\ 0 & 1 & 1 \\ 1 & 2 & 0 \end{array}\right\rbrack = \left\lbrack \be...
Equation 4.8.10: The first line is equation 4.8.5 ; the second is the inductive assumption. and equation 4.8.5 corresponds to\n\n\[ {\Delta }_{3}\left( A\right) = 1\underset{i = 1}{\underbrace{{\Delta }_{2}\left( \left\lbrack \begin{array}{ll} 1 & 1 \\ 2 & 0 \end{array}\right\rbrack \right) }} - 0\underset{i = 2}{\unde...
Yes
Theorem 4.8.4. If \( A \) and \( B \) are \( n \times n \) matrices, then
Proof. The serious case is the one in which \( A \) is invertible. If \( A \) is invertible, consider the function\n\n\[ f\left( B\right) = \frac{\det \left( {AB}\right) }{\det A}.\n\nAs you can readily check (Exercise 4.8.8), it has properties 1, 2, and 3, which characterize the determinant function. Since the determi...
No
Corollary 4.8.5. If a matrix \( A \) is invertible, then\n\n\[ \det {A}^{-1} = \frac{1}{\det A} \]
Proof. Just compute: \( \det A\det {A}^{-1} = \det \left( {A{A}^{-1}}\right) = \det I = 1 \) .
Yes
Theorem 4.8.8. For any \( n \times n \) matrix \( A \) , \[ \det A = \det {A}^{\top }\text{.} \]
Proof. First we will see that the theorem is true for elementary matrices. \[ \det {E}_{3} = - \det I = - 1 \] The equation \( D\left( {A}_{2}\right) = {\mu D}\left( {A}_{1}\right) \) in the proof of uniqueness can be rewritten All type 1 matrices are diagonal as \( \det E = \mu \det I = \mu \), where \( E \) is an ele...
Yes
Theorem 4.8.9 (Determinant of triangular matrix). If a matrix is triangular, then its determinant is the product of the entries along the diagonal.
Proof. We will prove the result for upper triangular matrices; the result for lower triangular matrices then follows from Theorem 4.8.8. The proof is by induction. Theorem 4.8.9 is clearly true for a \( 1 \times 1 \) triangular matrix (note that any \( 1 \times 1 \) matrix is triangular). If \( A \) is triangular of si...
Yes
Example 4.8.13 (Computing the determinant by permutations). Let \( n = 3 \), and let \( A \) be the matrix \( A = \left\lbrack \begin{array}{lll} 1 & 2 & 3 \\ 4 & 5 & 6 \\ 7 & 8 & 9 \end{array}\right\rbrack \) . There are six possible permutations of \( n = 3 \) numbers, so we have the following, where the labeling of ...
So \( \det A = {45} + {84} + {96} - {48} - {72} - {105} = 0 \) . Can you see why this determinant had to be \( 0{?}^{10}\bigtriangleup \)
Yes
Example 4.8.16 (Derivative of the determinant of \( 2 \times 2 \) matrices). The determinant function of \( 2 \times 2 \) matrices can be considered as the function det : \( {\mathbb{R}}^{4} \rightarrow \mathbb{R} \) given by\n\n\( \det \left( \begin{array}{l} a \\ b \\ c \\ d \end{array}\right) = {ad} - {bc},\; \) wit...
Proof. 1. By Theorem 4.8.12, the determinant is a polynomial in the entries of the matrix, hence certainly differentiable. (For instance, the formula \( {ad} - {bc} \) is a polynomial in the variables \( a, b, c, d \) .)\n\n2. Since (Proposition 1.7.14) \( \left\lbrack {\mathbf{D}\det \left( I\right) }\right\rbrack B \...
No
Corollary 4.8.17. If \( P \) is invertible, then for any matrix \( A \) we have\n\n\[ \operatorname{tr}\left( {{P}^{-1}{AP}}\right) = \operatorname{tr}A \]
Proof. This uses Theorem 4.8.15 and Theorem 4.8.6 (basis independence of the determinant):
No
Theorem 4.8.15\n\n\\[ \n\\operatorname{tr}\\left( {{P}^{-1}{AP}}\\right) \\;\\text{ 和 }\\; \\triangleq \\;\\left\\lbrack {\\mathbf{D}\\det \\left( I\\right) }\\right\\rbrack \\left( {{P}^{-1}{AP}}\\right) \n\\]\n
\n\\[ \n= \\;\\mathop{\\lim }\\limits_{{h \\rightarrow 0}}\\frac{\\det \\left( {I + h{P}^{-1}{AP}}\\right) - \\det I}{h} \n\\]\n\n\\[ \n= \\;\\mathop{\\lim }\\limits_{{h \\rightarrow 0}}\\frac{\\det \\left( {{P}^{-1}\\left( {P + {hAP}}\\right) }\\right) - \\det I}{h} \n\\]\n\n\\[ \n= \\;\\mathop{\\lim }\\limits_{{h \\r...
Yes
If \( A = \left\lbrack \begin{array}{ll} 0 & 1 \\ 1 & 1 \end{array}\right\rbrack \), then
\n\[ {\chi }_{A}\left( t\right) = \det \left( {\left\lbrack \begin{array}{ll} t & 0 \\ 0 & t \end{array}\right\rbrack - \left\lbrack \begin{array}{ll} 0 & 1 \\ 1 & 1 \end{array}\right\rbrack }\right) = \det \left\lbrack \begin{matrix} t & - 1 \\ - 1 & t - 1 \end{matrix}\right\rbrack = {t}^{2} - t - 1.\;\bigtriangleup \...
Yes
Theorem 4.8.20. Let \( A \) be a square matrix. The eigenvalues of \( A \) are the roots of \( {\chi }_{A} \) .
Proof. If \( \lambda \) is a root of \( {\chi }_{A} \), then \( \det \left( {{\lambda I} - A}\right) = 0 \), so (by Theorem 4.8.3 and the dimension formula) \( \ker \left( {{\lambda I} - A}\right) \neq \{ \overline{\mathbf{0}}\} \) . If \( \overline{\mathbf{v}} \in \ker \left( {{\lambda I} - A}\right) \) is a nonzero v...
Yes
Corollary 4.8.21. If \( A \) is a triangular matrix, the diagonal entries of \( A \) are the eigenvalues of \( A \), each appearing as many times as its multiplicity as a root of \( {\chi }_{A} \) .
The proof is the object of Exercise 4.8.12.
No
Corollary 4.8.22. If the roots of \( {\chi }_{A} \) are simple, then \( {\mathbb{C}}^{n} \) admits an eigenbasis for \( A \) .
Proof. If the roots are simple, there are \( n \) of them, and the corresponding eigenvectors are linearly independent by Theorem 2.7.7, providing an eigenbasis.
Yes
Theorem 4.8.23. For any vector \( \overrightarrow{\mathbf{w}} \), the polynomial \( p \) divides \( {\chi }_{A} \) .
The roots of \( p \) are eigenvalues, so they are also roots of \( {\chi }_{A} \) . The problem is that we need to show that a root of \( p \) cannot have higher multiplicity than it does as a root of \( {\chi }_{A} \) . This requires three intermediate statements, all of great interest in their own right.
No
Proposition 4.8.24. If \( A \) is an \( n \times n \) complex matrix, there exists an invertible matrix \( P \) such that \( {P}^{-1}{AP} \) is upper triangular. Equivalently, there is a basis \( {\overrightarrow{\mathbf{v}}}_{1},\ldots ,{\overrightarrow{\mathbf{v}}}_{n} \) such that the matrix of \( A \) in that basis...
Proof. The proof is by induction on \( n \) . It is obvious if \( n = 1 \), so suppose \( n \geq 2 \) and assume the result for all \( \left( {n - 1}\right) \times \left( {n - 1}\right) \) matrices.\n\nFind an eigenvector \( {\overrightarrow{\mathbf{v}}}_{1} \) with eigenvalue \( {\lambda }_{1} \) (which exists by the ...
Yes
Theorem 4.8.26. All square matrices are in the closure of the diagonalizable ones: for every complex square matrix \( A \) there is a sequence of complex diagonalizable matrices \( {A}_{i} \) that converges to \( A \) .
Proof. Suppose that \( B\overset{\text{ def }}{ = }{P}^{-1}{AP} \) is upper triangular, with entries \( {\lambda }_{1},\ldots ,{\lambda }_{n} \) on the diagonal. Choose sequences \( {\lambda }_{i, m} \) such that for all \( m \) the numbers \( {\lambda }_{1, m},\ldots ,{\lambda }_{n, m} \) are distinct, and such that \...
Yes
Theorem 4.8.27 (The Cayley-Hamilton theorem). If \( A \) is any square matrix, then \( {\chi }_{A}\left( A\right) = \left\lbrack 0\right\rbrack \) .
Algebraic proofs of this theorem are quite difficult, but with Theorem 4.8.26 it is easy, and its proof (with hints) is the object of Exercise 4.8.14.
No
Corollary 4.9.2. If \( S \) is an orthogonal matrix, then\n\n\[{\operatorname{vol}}_{n}S\left( A\right) = {\operatorname{vol}}_{n}A\]
Proof. Since \( {S}^{\top }S = I \) (margin note next to Definition 2.4.15) we have \( \left( {\det S}\right) \left( {\det {S}^{\top }}\right) = {\left( \det S\right) }^{2} = 1 \), so \( \left| {\det S}\right| = 1 \) .
Yes
Lemma 4.9.6. The sequence of pavings \( N \mapsto T\left( {{\mathcal{D}}_{N}\left( {\mathbb{R}}^{n}\right) }\right) \) is a nested partition.
Proof of Lemma 4.9.6. We must check the two conditions of Definition 4.7.3 of a nested partition. The first condition is that small paving pieces must fit inside big paving pieces: if we pave \( {\mathbb{R}}^{n} \) with blocks \( T\left( C\right) \), then if\n\n\[ {C}_{1} \in {\mathcal{D}}_{{N}_{1}}\left( {\mathbb{R}}^...
Yes
Lemma 4.9.7. If \( S, T : {\mathbb{R}}^{n} \rightarrow {\mathbb{R}}^{n} \) are linear transformations, then\n\n\[ \n{\operatorname{vol}}_{n}\left( {S \circ T}\right) \left( Q\right) = {\operatorname{vol}}_{n}S\left( Q\right) {\operatorname{vol}}_{n}T\left( Q\right) \n\]
Proof of Lemma 4.9.7. This follows from equation 4.9.20, substituting \( S \) for \( T \) and \( T\left( Q\right) \) for \( A \) :\n\n\[ \n{\operatorname{vol}}_{n}\left( {S \circ T}\right) \left( Q\right) = {\operatorname{vol}}_{n}S\left( {T\left( Q\right) }\right) = {\operatorname{vol}}_{n}S\left( Q\right) {\operatorn...
Yes
Theorem 4.9.8 (Linear change of variables). Let \( T : {\mathbb{R}}^{n} \rightarrow {\mathbb{R}}^{n} \) be an invertible linear transformation, and \( f : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) an integrable function. Then \( f \circ T \) is integrable, and
\[ {\int }_{{\mathbb{R}}^{n}}f\left( \mathbf{y}\right) \left| {{d}^{n}\mathbf{y}}\right| = \underset{\begin{matrix} \text{ corrects for } \\ \text{ stretching by }T \end{matrix}}{\underbrace{\left| \det T\right| }}{\int }_{{\mathbb{R}}^{n}}\underset{f\left( \mathbf{y}\right) }{\underbrace{f\left( {T\left( \mathbf{x}\ri...
Yes
Example 4.9.9 (Linear change of variables). The linear transformation given by \( T = \left\lbrack \begin{array}{ll} a & 0 \\ 0 & b \end{array}\right\rbrack \) transforms the unit disc into an ellipse, as shown in Figure 4.9.6. The area of the ellipse is then given by
Area of ellipse \( = {\int }_{\text{ellipse }}\left| {{d}^{2}\mathbf{y}}\right| = \underset{ab}{\underbrace{\left| \det \left\lbrack \begin{array}{ll} a & 0 \\ 0 & b \end{array}\right\rbrack \right| }}\underset{\pi = \text{area of unit disc }}{\underbrace{{\int }_{\text{disc }}\left| {{d}^{2}\mathbf{x}}\right| }} = \le...
Yes
To compute\n\n\[ \n{\int }_{0}^{\pi }\sin x{e}^{\cos x}{dx} \n\]
traditionally, one sets \( u = \cos x \), so that \( {du} = - \sin {xdx} \) . Then for \( x = 0 \) , we have \( u = \cos 0 = 1 \), and for \( x = \pi \), we have \( u = \cos \pi = - 1 \), so\n\n\[ \n{\int }_{0}^{\pi }\sin x{e}^{\cos x}{dx} = {\int }_{1}^{-1} - {e}^{u}{du} = {\int }_{-1}^{1}{e}^{u}{du} = e - \frac{1}{e}...
Yes
Consider the paraboloid of Figure 4.10.2, given by\n\n\\[ \nz = f\\left( \\begin{array}{l} x \\\\ y \\end{array}\\right) = \\left\\{ \\begin{array}{ll} {x}^{2} + {y}^{2} & \\text{ if }{x}^{2} + {y}^{2} \\leq {R}^{2} \\\\ 0 & \\text{ if }{x}^{2} + {y}^{2} > {R}^{2} \\end{array}\\right.\n\\]\n
Usually one would write the integral\n\n\\[ \n{\\int }_{{\\mathbb{R}}^{2}}f\\left( \\begin{array}{l} x \\\\ y \\end{array}\\right) \\left| {dxdy}\\right| \\;\\text{ as }\\;{\\int }_{{D}_{R}}\\left( {{x}^{2} + {y}^{2}}\\right) \\left| {dxdy}\\right|\n\\]\n\nwhere\n\n\\[ \n{D}_{R} = \\left\\{ {\\left. {\\left( \\begin{ar...
No
The lemniscate looks like a figure eight; the name comes from the Latin word for ribbon. We will compute the area of the right lobe \( A \) of the lemniscate given by the equation
\[ = {\int }_{-\pi /4}^{\pi /4}\frac{\cos {2\theta }}{2}{d\theta } = {\left\lbrack \frac{\sin {2\theta }}{4}\right\rbrack }_{-\pi /4}^{\pi /4} = \frac{1}{2}. \]
Yes
Proposition 4.10.7 (Change of variables for spherical coordinates). Let \( f \) be an integrable function defined on \( {\mathbb{R}}^{3} \), and let the spherical coordinate map \( S \) map a region \( B \) of \( \left( {r,\theta ,\varphi }\right) \) -space to a region \( A \) in \( \left( {x, y, z}\right) \) -space. S...
The \( {r}^{2}\cos \varphi \) corrects for distortion induced by the spherical coordinates map. Again, we postpone the justification for this formula.
No
Let's integrate the function \( z \) over the upper half of the unit ball, denoted \( A \):
\[ {\int }_{A}z\left| {dxdydz}\right| \text{becomes} \]\n\[ {\int }_{B}\underset{z}{\underbrace{\left( r\sin \varphi \right) }}\left( {{r}^{2}\cos \varphi }\right) \left| {drd\theta d\varphi }\right| = {\int }_{0}^{1}\left( {{\int }_{0}^{\pi /2}\left( {{\int }_{0}^{2\pi }{r}^{3}\sin \varphi \cos {\varphi d\theta }}\rig...
Yes
Proposition 4.10.10 (Change of variables for cylindrical coordinates). Let \( f \) be an integrable function defined on \( {\mathbb{R}}^{3} \), and suppose that the cylindrical coordinate map \( C \) maps a region \( B \subset \left( {0,\infty }\right) \times \lbrack 0,{2\pi }) \times \mathbb{R} \) of \( \left( {r,\the...
\[ {\int }_{A}f\left( \begin{array}{l} x \\ y \\ z \end{array}\right) \left| {dxdydz}\right| = {\int }_{B}f\left( \begin{matrix} r\cos \theta \\ r\sin \theta \\ z \end{matrix}\right) r\left| {drd\theta dz}\right| \]
Yes
Let us integrate \( \left( {{x}^{2} + {y}^{2}}\right) z \) over the region \( A \subset {\mathbb{R}}^{3} \) that is the part of the cone \( {z}^{2} \geq {x}^{2} + {y}^{2} \) where \( 0 \leq z \leq 1 \) (see Figure 4.10.6).
\[ {\int }_{A}({x}^{2} + {y}^{2})z\;|{dx}\;{dy}\;{dz}| = {\int }_{B}{r}^{2}z(\underset{ = 1}{\underbrace{{\cos }^{2}\theta + {\sin }^{2}\theta }})\;r\;|{dr}\;{d\theta }\;{dz}| = {\int }_{B}({r}^{2}z)\;r\;|{dr}\;{d\theta }\;{dz}| \] \[ = {\int }_{0}^{2\pi }\left( {{\int }_{0}^{1}\left( {{\int }_{r}^{1}{r}^{3}{zdz}}\righ...
Yes
Theorem 4.10.12 (Change of variables formula). Let \( X \) be a compact subset of \( {\mathbb{R}}^{n} \) with boundary \( \partial X \) of volume 0 ; let \( U \subset {\mathbb{R}}^{n} \) be an open set containing \( X \) . Let \( \Phi : U \rightarrow {\mathbb{R}}^{n} \) be a \( {C}^{1} \) mapping that is injective on \...
The theorem is proved in Appendix A19.
Yes
Consider the ratio of equation 4.10.29 in the case of polar coordinates, when \( \Phi = P \) . If a rectangle \( C \) in the \( \left( {r,\theta }\right) \) plane, containing the point \( \left( \begin{matrix} {r}_{0} \\ {\theta }_{0} \end{matrix}\right) \), has sides of length \( {\Delta r} \) and \( {\Delta \theta } ...
\[ {\int }_{Y}f\left| {{d}^{n}\mathbf{y}}\right| = {\int }_{X}\left( {f \circ P}\right) r\left| {drd\theta }\right| \] where \( r \) is the ratio of the volumes of infinitesimal paving blocks.
Yes
Suppose you wish to find the area of the region \( X \subset {\mathbb{R}}^{2} \) given by\n\n\[ 1 \leq {xy} \leq 2\text{ and }{x}^{2} \leq y \leq 2{x}^{2}. \]\n\nWhat change of variables is appropriate?
First we draw the hyperbolas given by the equalities \( 1 = {xy} \) and \( {xy} = 2 \), and the parabolas given by \( {x}^{2} = y \) and \( y = 2{x}^{2} \), as shown in Figure 4.10.9. This figure suggests that setting \( u = {xy} \) would be one good choice; the equation \( {x}^{2} \leq y \leq 2{x}^{2} \) suggests that...
No
Example 4.10.19 (A less standard change of variables). The region \( T \) defined by\n\n\[ \n{\\left( \\frac{x}{1 - z}\\right) }^{2} + {\\left( \\frac{y}{1 + z}\\right) }^{2} \\leq 1,\\; - 1 < z < 1 \n\]
looks like the curvy-sided tetrahedron pictured in Figure 4.10.10; we will compute its volume. Notice that horizontal slices of \( T \) are ellipses, so we will use \
No
Theorem 4.11.2 (Convergence for Riemann integrals). Let \( k \mapsto {f}_{k} \) be a sequence of integrable functions \( {\mathbb{R}}^{n} \rightarrow \mathbb{R} \), all with support in a fixed ball \( B \subset {\mathbb{R}}^{n} \), and converging uniformly to a function \( f \) . Then \( f \) is integrable, and \[ \mat...
Proof. Choose \( \epsilon > 0 \) and \( K \) so large that \( \mathop{\sup }\limits_{{\mathbf{x} \in {\mathbb{R}}^{n}}}\left| {f\left( \mathbf{x}\right) - {f}_{k}\left( \mathbf{x}\right) }\right| < \epsilon \) when \( k > K \) . Then when \( k > K \), we have, for any \( N \), \[ {L}_{N}\left( f\right) > {L}_{N}\left( ...
Yes
the mass of the integral is contained in a square 1 high and 1 wide. As \( k \rightarrow \infty \) this mass drifts off to infinity and gets lost:
\n\[ \mathop{\lim }\limits_{{k \rightarrow \infty }}{\int }_{0}^{\infty }{f}_{k}\left( x\right) {dx} = 1,\;\text{ but }{\int }_{0}^{\infty }\mathop{\lim }\limits_{{k \rightarrow \infty }}{f}_{k}\left( x\right) {dx} = {\int }_{0}^{\infty }{0dx} = 0. \]
Yes
Theorem 4.11.4 (Dominated convergence for Riemann integrals). Let \( {f}_{k} : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) be a sequence of \( R \) -integrable functions. Suppose there exists \( R \) such that all \( {f}_{k} \) have their support in \( {B}_{R}\left( \mathbf{0}\right) \) and satisfy \( \left| {f}_{k}\rig...
\[ \mathop{\lim }\limits_{{k \rightarrow \infty }}{\int }_{{\mathbb{R}}^{n}}{f}_{k}\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| = {\int }_{{\mathbb{R}}^{n}}f\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| \]
Yes
Theorem 4.11.7. Let \( k \mapsto {f}_{k}, k \mapsto {g}_{k} \) be two sequences of \( R \) -integrable functions such that\n\n\[ \mathop{\sum }\limits_{{k = 1}}^{\infty }{\int }_{{\mathbb{R}}^{n}}\left| {{f}_{k}\left( \mathbf{x}\right) }\right| \left| {{d}^{n}\mathbf{x}}\right| < \infty ,\;\mathop{\sum }\limits_{{k = 1...
Proof of Theorem 4.11.7. Set \( {h}_{k} = {f}_{k} - {g}_{k} \), and \( {H}_{l} = \mathop{\sum }\limits_{{k = 1}}^{l}{h}_{k} \) . To prove equation 4.11.18 we need to show that\n\n\[ \mathop{\lim }\limits_{{l \rightarrow \infty }}{\int }_{{\mathbb{R}}^{n}}{H}_{l}\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right|...
Yes
Proposition 4.11.9. If \( f \) is R-integrable, then it is L-integrable, and its Lebesgue integral equals its Riemann integral.
Proof. Just take \( {f}_{1} = f \), and set \( {f}_{k} = 0 \) for \( k = 2,3,\ldots \) Clearly \( \sum {f}_{k} = f \) everywhere, and inequality 4.11.19 is satisfied:\n\n\[ \mathop{\sum }\limits_{{k = 1}}^{\infty }\int \left| {{f}_{k}\left( \mathbf{x}\right) }\right| \left| {{d}^{n}\mathbf{x}}\right| = \int \left| {{f}...
Yes
Example 4.11.10 (L-integrable function that is not R-integrable). Let \( f \) be the indicator function of the rationals, i.e., \[ f\left( x\right) = \left\{ \begin{array}{ll} 1 & \text{ if }x \in \mathbb{Q} \\ 0 & \text{ otherwise. } \end{array}\right. \]
This function equals 0 almost everywhere, so it is Lebesgue integrable with integral \( 0.\;\bigtriangleup \)
Yes
Example 4.11.12 (An unbounded, integrable function). The function \( f\left( x\right) = {\mathbf{1}}_{\left\lbrack 0,1\right\rbrack }\left( x\right) \ln x \) is L-integrable, even though it isn’t bounded.
As shown in Figure 4.11.4, it can be written as a sum of bounded functions:\n\n\[ f\left( x\right) \underset{L}{ = }\mathop{\sum }\limits_{{i = 0}}^{\infty }{f}_{i}\left( x\right) ,\;\text{ where }{f}_{i} = \left( {{\mathbf{1}}_{\left( {2}^{-\left( {i + 1}\right) },{2}^{-i}\right\rbrack }\left( x\right) }\right) \ln x....
Yes
Example 4.11.13 (A function that is not Lebesgue integrable). Some improper one-dimensional integrals do not correspond to Lebesgue-integrable functions: integrals whose existence depends on cancellations, like\n\n\[ \n{\int }_{0}^{\infty }\frac{\sin x}{x}{dx} \n\]
As you may recall from one-variable calculus, this improper integral is defined to be\n\n\[ \n\mathop{\lim }\limits_{{A \rightarrow \infty }}{\int }_{0}^{A}\frac{\sin x}{x}{dx} \n\]\n\nand we can show that the limit exists, for instance, by saying that the series\n\n\[ \n\mathop{\sum }\limits_{{k = 0}}^{\infty }{\int }...
Yes
Proposition 4.11.14 (The Lebesgue integral is linear). If \( f \) and \( g \) are L-integrable and \( a, b \) are constants, then \( {af} + {bg} \) is L-integrable and\n\n\[{\int }_{{\mathbb{R}}^{n}}\left( {{af} + {bg}}\right) \left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| = a{\int }_{{\mathbb{R}}^{n}}f\lef...
Proof. If \( f = \sum {f}_{k} \) and \( g = \sum {g}_{k} \), then \( {af} + {bg} = \sum \left( {a{f}_{k} + b{g}_{k}}\right) \) . Indeed, the series \( \sum \left( {a{f}_{k} + b{g}_{k}}\right) \) will converge except on the union of the sets where the series \( \mathop{\sum }\limits_{k}\left| {f\left( \mathbf{x}\right) ...
Yes
Proposition 4.11.15. If \( f \) is L-integrable on \( {\mathbb{R}}^{n} \), and \( g \) is \( R \) -integrable on \( {\mathbb{R}}^{n} \), then \( {fg} \) is L-integrable.
Proof. Since \( f \) is L-integrable, we can set \( f = \mathop{\sum }\limits_{k}{f}_{k} \), where the functions \( {f}_{k} \) are R-integrable and\n\n\[ \mathop{\sum }\limits_{k}{\int }_{{\mathbb{R}}^{n}}\left| {{f}_{k}\left( \mathbf{x}\right) }\right| \begin{Vmatrix}{{d}^{n}\mathbf{x}}\end{Vmatrix} < \infty \]\n\nWe ...
Yes
Proposition 4.11.16. If \( f \) and \( g \) are \( L \) -integrable and \( f\underset{L}{ \leq }g \), then \[ {\int }_{{\mathbb{R}}^{n}}f\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| \leq {\int }_{{\mathbb{R}}^{n}}g\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| \]
Proof. By Proposition 4.11.14, the statement is equivalent to saying that \( 0 \leq g - f \) implies \( 0 \leq {\int }_{{\mathbb{R}}^{n}}\left( {g - f}\right) \left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| \) . Replace \( g - f \) by \( f \) ; the statement becomes \( 0 \leq f \Rightarrow 0 \leq {\int }_{{\...
Yes
Theorem 4.11.18. (Monotone convergence theorem: Lebesgue integrals). 1. Let \( 0 \leq {f}_{1}\underset{L}{ \leq }{f}_{2}\underset{L}{ \leq }\cdots \) be a sequence of L-integrable nonnegative functions. If \[ \mathop{\sup }\limits_{k}{\int }_{{\mathbb{R}}^{n}}{f}_{k}\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\r...
Proof. 1. Apply Theorem 4.11.17 to the sup, rewritten as the series \[ f = {f}_{1} + \left( {{f}_{2} - {f}_{1}}\right) + \cdots + \left( {{f}_{k} - {f}_{k - 1}}\right) + \cdots = {g}_{1} + {g}_{2} + \cdots + {g}_{k} + \cdots . \] 2. If \( f \) were L-integrable, then by Proposition 4.11.16 and \( f \geq {f}_{k} \) , \[...
Yes
Theorem 4.11.19 (Dominated convergence theorem: Lebesgue integrals). Let \( k \mapsto {f}_{k} \) be a sequence of L-integrable functions that converges pointwise almost everywhere to some function \( f \) . Suppose there is an L-integrable function \( F : {\mathbb{R}}^{n} \rightarrow \mathbb{R} \) such that \( \left| {...
\[ {\int }_{{\mathbb{R}}^{n}}f\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| = \mathop{\lim }\limits_{{k \rightarrow \infty }}{\int }_{{\mathbb{R}}^{n}}{f}_{k}\left( \mathbf{x}\right) \left| {{d}^{n}\mathbf{x}}\right| \]
Yes
Theorem 4.11.22 (Differentiating under the integral sign). Let \( f\left( {t,\mathbf{x}}\right) : {\mathbb{R}}^{n + 1} \rightarrow \mathbb{R} \) be a function such that for each fixed \( t \), the integral\n\n\[ F\left( t\right) = {\int }_{{\mathbb{R}}^{n}}f\left( {t,\mathbf{x}}\right) \left| {{d}^{n}\mathbf{x}}\right|...
Proof. Just compute\n\n\[ {F}^{\prime }\left( t\right) = \mathop{\lim }\limits_{{h \rightarrow 0}}\frac{F\left( {t + h}\right) - F\left( t\right) }{h} = \mathop{\lim }\limits_{{h \rightarrow 0}}{\int }_{{\mathbb{R}}^{n}}\frac{f\left( {t + h,\mathbf{x}}\right) - f\left( {t,\mathbf{x}}\right) }{h}\left| {{d}^{n}\mathbf{x...
Yes
Let\n\n\\[ \n{f}_{a}\left( x\right) = {e}^{-a{x}^{2}},\;\text{ so that }\;{\widehat{f}}_{a}\left( \xi \right) = {\int }_{\mathbb{R}}{e}^{-a{x}^{2}}{e}^{2\pi i\xi x}{dx}. \n\\]
We can't compute this Fourier transform directly, but equation 4.11.85 (or Theorem 4.11.22) gives\n\n\\[ \n{\widehat{{f}_{a}}}^{\prime }\left( \xi \right) = {\pi i}{\int }_{-\infty }^{\infty }{e}^{2\pi ix\xi }\left( {{2x}{e}^{-a{x}^{2}}}\right) {dx}. \n\\]\n\nThis can be integrated by parts (justified by Exercise 4.11....
Yes
The Fourier transform of both sides of the differential equation\n\n\\[ \n{a}_{p}{D}^{p}f + \cdots + {a}_{0}f = g \n\\]
is\n\n\\[ \n\\underset{\\text{product of }\\widehat{f}\\text{ and a polynomial }}{\\underbrace{\\left( {{a}_{p}{\\left( -2\\pi i\\xi \\right) }^{p} + {a}_{p - 1}{\\left( -2\\pi i\\xi \\right) }^{p - 1} + \\cdots + {a}_{0}}\\right) \\widehat{f}}} = \\widehat{g}, \n\\]\n\nwhich gives\n\n\\[ \n\\widehat{f} = \\frac{\\wide...
Yes
Proposition 5.1.1 (Volume of a \( k \) -parallelogram in \( {\mathbb{R}}^{k} \) ). Let \( {\overrightarrow{\mathbf{v}}}_{1},\ldots ,{\overrightarrow{\mathbf{v}}}_{k} \) be \( k \) vectors in \( {\mathbb{R}}^{k} \), so that \( T = \left\lbrack {{\overrightarrow{\mathbf{v}}}_{1},\ldots ,{\overrightarrow{\mathbf{v}}}_{k}}...
Proof of Proposition 5.1.1:\n\nRecall that if \( A \) and \( B \) are \( n \times n \) Proof. \( \sqrt{\det \left( {{T}^{\top }T}\right) } = \sqrt{\left( {\det {T}^{\top }}\right) \left( {\det T}\right) } = \sqrt{{\left( \det T\right) }^{2}} = \left| {\det T}\right| \) matrices, then\n\n\[ \n\det A\det B = \det \left( ...
Yes
Example 5.1.2 (Volume of two-dimensional and three-dimensional parallelograms). When \( k = 2 \), we have
\[ \det \left( {{T}^{\top }T}\right) = \det \left( {\left\lbrack \begin{matrix} {\overrightarrow{\mathbf{v}}}_{1}^{\top } \\ {\overrightarrow{\mathbf{v}}}_{2}^{\top } \end{matrix}\right\rbrack \left\lbrack \begin{matrix} {\overrightarrow{\mathbf{v}}}_{1} & {\overrightarrow{\mathbf{v}}}_{2} \end{matrix}\right\rbrack }\r...
Yes
Example 5.1.4 (Volume of a 3-parallelogram in \( {\mathbb{R}}^{4} \) ). Let \( P \) be the 3-parallelogram \( P \) in \( {\mathbb{R}}^{4} \) spanned by \( {\overrightarrow{\mathbf{v}}}_{1} = \left\lbrack \begin{array}{l} 1 \\ 0 \\ 0 \\ 1 \end{array}\right\rbrack ,{\overrightarrow{\mathbf{v}}}_{2} = \left\lbrack \begin{...
Set \( T = \left\lbrack {{\overrightarrow{\mathbf{v}}}_{1},{\overrightarrow{\mathbf{v}}}_{2},{\overrightarrow{\mathbf{v}}}_{3}}\right\rbrack \) ; then \[ {T}^{\top }T = \left\lbrack \begin{array}{lll} 2 & 1 & 1 \\ 1 & 2 & 1 \\ 1 & 1 & 2 \end{array}\right\rbrack \;\text{ and }\;\det \left( {{T}^{\top }T}\right) = 4,\;\t...
Yes
Example 5.2.4 (Parametrization of a cone). The subset of \( {\mathbb{R}}^{3} \) of equation \( {x}^{2} + {y}^{2} - {z}^{2} = 0 \), shown in Figure 5.2.1, is not a manifold in the neighborhood of the vertex, which is at the origin. However, the subset\n\n\[ M = \\left\\{ {\\left. \\left( \\begin{array}{l} x \\\\ y \\\\ ...
In the language of Definition 5.2.3 we have \( U = \\left\\lbrack {0,1}\\right\\rbrack \\times \\left\\lbrack {0,{2\\pi }}\\right\\rbrack \) . We will set \( X = \\partial U \), so that \( U - X = \\left( {0,1}\\right) \\times \\left( {0,{2\\pi }}\\right) \), and \( X \) consists of the four line segments \( \\left( \\...
No
Consider the surface in Figure 5.2.3, which is obtained by rotating the curve of equation \( {\left( 1 - x\right) }^{3} = {z}^{2} \) in the \( \left( {x, z}\right) \) -plane around the \( z \) -axis. This surface has the equation \( {\left( 1 - \sqrt{{x}^{2} + {y}^{2}}\right) }^{3} = {z}^{2} \) . The curve can be param...
Figure 5.2.3 shows the image of this parametrization for \( 0 \leq \theta \leq {3\pi }/2 \) and \( \left| t\right| \leq 1 \) . It can be guessed from the picture, and proved from the formula, that the points in \( \left\lbrack {-1,1}\right\rbrack \times \left\lbrack {0,{3\pi }/2}\right\rbrack \) where \( t = \pm 1 \) a...
Yes
Let \( {\gamma }_{1} \) and \( {\gamma }_{2} \) be two parametrizations of \( {S}^{2} \) by spherical coordinates, but with different poles. Call \( {P}_{1},{P}_{1}^{\prime } \) the poles for \( {\gamma }_{1} \) and \( {P}_{2},{P}_{2}^{\prime } \) the poles for \( {\gamma }_{2} \) . Then \( {\gamma }_{2}^{-1} \circ {\g...
Indeed, some single point in the domain of \( {\gamma }_{1} \) maps to \( {P}_{2} \cdot {}^{2} \) But as shown in Figure 5.2.4, \( {\gamma }_{2} \) maps a whole segment to \( {P}_{2} \), so that \( {\gamma }_{2}^{-1} \circ {\gamma }_{1} \) \
No
Theorem 5.2.11. Both \( {U}_{1}^{\mathrm{{OK}}} = {U}_{1} - \left( {{X}_{1} \cup {Y}_{2}}\right) \) and \( {U}_{2}^{\mathrm{{OK}}} = {U}_{2} - \left( {{X}_{2} \cup {Y}_{1}}\right) \) are open subsets of \( {\mathbb{R}}^{k} \) with boundaries of \( k \) -dimensional volume 0, and\n\n\[ \Phi \overset{\text{ def }}{ = }{\...
Proof. By Proposition 3.2.11 we have\n\n\[ \left\lbrack {\mathbf{D}\Phi \left( \mathbf{x}\right) }\right\rbrack = \left\lbrack {\mathbf{D}{\gamma }_{2}^{-1}\left( {{\gamma }_{1}\left( \mathbf{x}\right) }\right) }\right\rbrack \left\lbrack {\mathbf{D}{\gamma }_{1}\left( \mathbf{x}\right) }\right\rbrack \] \nsince both \...
Yes
Proposition 5.3.3 (Integral independent of parametrization). Let \( M \) be a \( k \) -dimensional manifold in \( {\mathbb{R}}^{n} \) and \( f : M \rightarrow \mathbb{R} \) a function. If \( U \) and \( V \) are subsets of \( {\mathbb{R}}^{k} \) and \( {\gamma }_{1} : U \rightarrow M,{\gamma }_{2} : V \rightarrow M \) ...
Proof. Define \( \Phi = {\gamma }_{2}^{-1} \circ {\gamma }_{1} : {U}^{\mathrm{{OK}}} \rightarrow {V}^{\mathrm{{OK}}} \) to be the \
No
Corollary 5.3.4. Every point of a \( k \) -dimensional manifold \( M \subset {\mathbb{R}}^{n} \) has a neighborhood with finite \( k \) -dimensional volume.
Proof. This follows from Theorem 5.2.6 and Proposition 5.3.3 (and Theorem 4.3.6, which says that the integral, hence volume, exists).
Yes