Product Integration and Change of Variables Practice Quiz (30 questions)

All questions and answers

1. How is the product σ-algebra on E1×E2E_1 \times E_2 defined?

The σ-algebra generated by all subsets A×BA \times B of the product space
The σ-algebra generated by disjoint measurable rectangles A×BA \times B
The σ-algebra generated by all measurable rectangles A×BA \times B, where A∈E1A \in \mathcal{E}_1 and B∈E2B \in \mathcal{E}_2
The collection of all rectangles A×BA \times B, where A∈E1A \in \mathcal{E}_1 and B∈E2B \in \mathcal{E}_2

The σ-algebra generated by all measurable rectangles $$A \times B$$, where $$A \in \mathcal{E}_1$$ and $$B \in \mathcal{E}_2$$

Explanation

The product σ-algebra is the σ-algebra generated by measurable rectangles, so it includes the closure under σ-algebra operations rather than just the rectangles themselves. The other choices respectively omit that closure, allow arbitrary subsets, or add an unnecessary disjointness condition.

2. Suppose C∈E1⊗E2C \in \mathcal{E}_1 \otimes \mathcal{E}_2. What can be concluded about its sections for each fixed x∈E1x \in E_1 and y∈E2y \in E_2?

Each CxC_x belongs to E1\mathcal{E}_1, and each CyC^y belongs to E2\mathcal{E}_2
Each CxC_x belongs to E2\mathcal{E}_2, and each CyC^y belongs to E1\mathcal{E}_1
The sections belong to their respective σ-algebras for almost every fixed coordinate
The sections belong to their respective σ-algebras only when CC is a rectangle

Each $$C_x$$ belongs to $$\mathcal{E}_2$$, and each $$C^y$$ belongs to $$\mathcal{E}_1$$

Explanation

For a product-measurable set, every fixed-coordinate section is measurable in the σ-algebra of the remaining coordinate: Cx∈E2C_x \in \mathcal{E}_2 and Cy∈E1C^y \in \mathcal{E}_1. Reversing those spaces, weakening the result to almost every coordinate, or restricting it to rectangles misstates the section property.

3. A product-measurable function f:E1×E2→Gf:E_1 \times E_2 \to G is fixed at coordinates x0∈E1x_0 \in E_1 and y0∈E2y_0 \in E_2. Which measurability conclusion applies to its resulting one-variable functions?

fx0f_{x_0} is E2\mathcal{E}_2-measurable and fy0f^{y_0} is E1\mathcal{E}_1-measurable
fx0f_{x_0} is E1\mathcal{E}_1-measurable and fy0f^{y_0} is E2\mathcal{E}_2-measurable
Only fx0f_{x_0} is guaranteed to be measurable, with respect to E2\mathcal{E}_2
Both fx0f_{x_0} and fy0f^{y_0} are measurable on the product space

$$f_{x_0}$$ is $$\mathcal{E}_2$$-measurable and $$f^{y_0}$$ is $$\mathcal{E}_1$$-measurable

Explanation

Fixing one coordinate leaves a measurable function of the other coordinate, so fx0f_{x_0} is measurable with respect to E2\mathcal{E}_2 and fy0f^{y_0} with respect to E1\mathcal{E}_1. The reversed structures swap the domains, joint measurability concerns a different function, and the conclusion covers both sections.

4. A measure μ\mu on (E,E)(E,\mathcal{E}) has measurable sets XnX_n that increase to EE and satisfy μ(Xn)<∞\mu(X_n)<\infty for every nn. How is μ\mu classified?

σ-finite
Locally finite
A probability measure
Finite

σ-finite

Explanation

A measure is σ-finite when an increasing sequence of measurable sets covers the space and each set has finite measure. Finiteness requires the whole space to have finite measure, while local finiteness and probability measure impose different conditions.

5. A measurable space is covered by measurable sets XnX_n increasing to EE, and every XnX_n has finite μ\mu-measure. Which property of μ\mu does this establish?

Normalization of μ\mu as a probability measure
Continuity of μ\mu
σ-finiteness
Finiteness of μ(E)\mu(E)

σ-finiteness

Explanation

An increasing measurable cover by sets of finite measure is precisely the criterion for σ-finiteness. It does not imply that the entire space has finite measure, nor does it establish continuity or probability normalization.

6. Let μ and ν be σ-finite measures and suppose μ(A)=0 while ν(B)=∞. What value does their product measure assign to the rectangle A × B under the stated convention?

0
μ(A) + ν(B)
∞
Not determined by the rectangle rule

0

Explanation

The rectangle rule gives μ⊗ν(A × B)=μ(A)ν(B), and the specified convention sets 0 × ∞ to 0. Assigning infinity ignores that convention, while adding the marginal measures is not the product-measure rule.

7. For a measurable set C in a product space, which expression computes its product measure by integrating the measures of its vertical sections?

∫E1μ(Cy) μ(dx)\int_{E_1} \mu(C_y)\,\mu(dx)
∫E2ν(Cx) ν(dy)\int_{E_2} \nu(C_x)\,\nu(dy)
∫E1ν(Cx) μ(dx)\int_{E_1} \nu(C_x)\,\mu(dx)
μ(E1)ν(E2)\mu(E_1)\nu(E_2)

$$\int_{E_1} \nu(C_x)\,\mu(dx)$$

Explanation

The product measure can be computed by integrating the ν-measure of each x-section against μ on E₁; equivalently, one can integrate the μ-measures of y-sections against ν on E₂. The other expressions mismatch the section and integrating measure or use the whole spaces rather than C.

8. A measure m on E₁ × E₂ is defined by integrating ω(Cₓ) with respect to μ over E₁. Which expression gives m(C) for a measurable set C?

∫E1ω(Cx) μ(dx)\int_{E_1}\omega(C_x)\,\mu(dx)
∫E2μ(Cy) ω(dy)\int_{E_2}\mu(C_y)\,\omega(dy)
∫E1ω(Cx) ω(dx)\int_{E_1}\omega(C_x)\,\omega(dx)
∫E1μ(C) μ(dx)\int_{E_1}\mu(C)\,\mu(dx)

$$\int_{E_1}\omega(C_x)\,\mu(dx)$$

Explanation

By definition, m(C) is the integral over E₁ of the ω-measure of the x-section, taken with respect to μ. The reversed section formula changes the defining setup, and the other alternatives use the wrong integrating measure or integrand.

9. Suppose μ(A)=3 and ω(B)=4 for measurable sets A and B. What is m(A × B) for the measure m defined from μ and ω?

7
12
24
μ(B)ω(A)\mu(B)\omega(A)

12

Explanation

For a measurable rectangle, m(A × B)=μ(A)ω(B), so 3 × 4 = 12. Adding gives 7, reversing the assignments does not follow the rectangle rule, and the formula has no extra factor.

10. Three random variables X₁, X₂, and X₃ are claimed to be independent. Which condition must hold for every measurable choice B₁, B₂, B₃ and every subset J of {1,2,3}?

P(⋂k∈J{Xk∈Bk})=∏k∈JP(Xk∈Bk)P\left(\bigcap_{k\in J}\{X_k\in B_k\}\right)=\prod_{k\in J}P(X_k\in B_k)
P(⋂k∈J{Xk∈Bk})=∏k∈JP(Xk∈Bk)for pairs onlyP\left(\bigcap_{k\in J}\{X_k\in B_k\}\right)=\prod_{k\in J}P(X_k\in B_k)\quad\text{for pairs only}
P(⋃k∈J{Xk∈Bk})=∏k∈JP(Xk∈Bk)P\left(\bigcup_{k\in J}\{X_k\in B_k\}\right)=\prod_{k\in J}P(X_k\in B_k)
P(⋂k=13{Xk∈Bk})=∏k=13P(Xk∈Bk)P\left(\bigcap_{k=1}^{3}\{X_k\in B_k\}\right)=\prod_{k=1}^{3}P(X_k\in B_k)

$$P\left(\bigcap_{k\in J}\{X_k\in B_k\}\right)=\prod_{k\in J}P(X_k\in B_k)$$

Explanation

Independence requires the stated factorization for every subset of variables and every choice of measurable sets, not merely for the full collection or pairs. Replacing intersections with unions changes the defining condition.

11. A pair (X,Y) has joint law equal to the product of its two marginal laws. What does this imply about X and Y?

X and Y have identical marginal distributions
X and Y are dependent unless their marginals are identical
X and Y are merely uncorrelated
X and Y are independent

X and Y are independent

Explanation

Two random variables are independent if and only if their joint law is the product of their marginal laws. Equality of the marginals is a different property, and uncorrelatedness alone does not express this joint-law condition.

12. A nonnegative simple function on a product space equals 33 on a measurable set of measure 22 and 55 on a disjoint measurable set of measure 44, and is zero elsewhere. What is its integral?

88
2626
5252
2222

$$26$$

Explanation

For a nonnegative simple function, multiply each value by the measure of its set and add: 3×2+5×4=263\times2+5\times4=26. The value 88 omits set measures, 2222 reuses the smaller set's measure, and 5252 multiplies rather than adds the contributions.

13. A measurable function ff satisfies ∫∣f∣ dμ=7\int |f|\,d\mu=7. Which conclusion follows about its membership in L1(μ)L^1(\mu)?

Its membership cannot be determined without knowing ∫f dμ\int f\,d\mu
It belongs to L1(μ)L^1(\mu) because the absolute-value integral is finite
It belongs to L1(μ)L^1(\mu) only if ff is bounded
It does not belong to L1(μ)L^1(\mu) because the integral is nonzero

It belongs to $$L^1(\mu)$$ because the absolute-value integral is finite

Explanation

A measurable function belongs to L1(μ)L^1(\mu) exactly when ∫∣f∣ dμ<+∞\int |f|\,d\mu<+\infty, and the given value is finite. Requiring a zero integral confuses integrability with cancellation, while boundedness or the signed integral is not the criterion.

14. For Borel sets A1,A2,A3A_1,A_2,A_3 with Lebesgue measures 2,4,52,4,5, respectively, what is λ⊗3(A1×A2×A3)\lambda^{\otimes 3}(A_1\times A_2\times A_3)?

55
1111
4040
1010

$$40$$

Explanation

The product measure of a measurable rectangle is the product of the coordinate measures: 2×4×5=402\times4\times5=40. The value 1111 comes from adding, 1010 omits one factor, and 55 uses just the largest measure.

15. A set in Rd\mathbb{R}^d is measurable for Lebesgue measure. Which description captures how Lebesgue measure is constructed?

It is defined on the Borel sigma-algebra alone, with no completion
It is outer measure on all subsets, without a measurable-sigma-algebra restriction
It is the restriction of outer measure to its measurable sigma-algebra, a completion containing the Borel sigma-algebra
It is a restriction to a sigma-algebra contained within the Borel sigma-algebra

It is the restriction of outer measure to its measurable sigma-algebra, a completion containing the Borel sigma-algebra

Explanation

Lebesgue measure on Rd\mathbb{R}^d is outer measure restricted to its measurable sigma-algebra, which contains the Borel sigma-algebra and completes it with respect to Lebesgue measure. Borel sets alone omit the completion, while the other descriptions misstate the restriction or containment.

16. A measure on the Borel sigma-algebra of Rd\mathbb{R}^d assigns every d-cell its ordinary volume. What does the uniqueness property imply?

It must be Lebesgue measure on the Borel sigma-algebra
It must be the product measure on every larger sigma-algebra
It is determined only after extending the measure to all subsets
It may be any measure agreeing with volume on one selected cell

It must be Lebesgue measure on the Borel sigma-algebra

Explanation

Lebesgue measure is the unique measure on the Borel sigma-algebra of Rd\mathbb{R}^d that assigns each d-cell its volume. Agreement on just one cell is insufficient, and the uniqueness statement concerns the Borel sigma-algebra rather than all subsets or every larger sigma-algebra.

17. A map between open subsets of Rd\mathbb{R}^d is bijective, and both the map and its inverse are differentiable with continuous differentials. How is this map classified?

A continuous bijection
A C1-diffeomorphism
A measure-preserving map
A differentiable immersion

A C1-diffeomorphism

Explanation

A C1-diffeomorphism is bijective, with both it and its inverse differentiable and having continuous differentials. Continuity alone, a constant Jacobian, and differentiability without the inverse condition do not give this classification.

18. Let φ:V→U\varphi:V\to U be a C1-diffeomorphism and let f:U→Rf:U\to\mathbb{R} be measurable. If ∣Jφ∣(f∘φ)|J_\varphi|(f\circ\varphi) is integrable on VV, what follows about ff?

ff is bounded on UU
ff is integrable on UU
ff is continuous on UU
ff has integral zero on UU

$$f$$ is integrable on $$U$$

Explanation

The integrability criterion is an if-and-only-if: integrability of the Jacobian-weighted composition on VV implies integrability of ff on UU. Boundedness, continuity, and a zero integral do not follow from this criterion.

19. For a C1-diffeomorphism φ:V→U\varphi:V\to U and an integrable function ff on UU, which expression gives its integral over UU using coordinates in VV?

∫V∣Jφ(x)∣f(x) ϑ(dx)\int_V |J_\varphi(x)|f(x)\,\vartheta(dx)
∫V∣Jφ(x)∣f(φ(x)) ϑ(dx)\int_V |J_\varphi(x)|f(\varphi(x))\,\vartheta(dx)
∫VJφ(x)f(φ(x)) ϑ(dx)\int_V J_\varphi(x)f(\varphi(x))\,\vartheta(dx)
∫Vf(φ(x))∣Jφ(x)∣ ϑ(dx)\int_V \frac{f(\varphi(x))}{|J_\varphi(x)|}\,\vartheta(dx)

$$\int_V |J_\varphi(x)|f(\varphi(x))\,\vartheta(dx)$$

Explanation

The change-of-variables formula weights the composed function by the absolute Jacobian determinant. Omitting the absolute value, failing to compose with the map, or using its reciprocal gives a different expression.

20. A C1-diffeomorphism φ:V→U\varphi:V\to U and a measurable function f:U→Rf:U\to\mathbb{R} are given. Which condition is equivalent to integrability of ff on UU?

ff is measurable on UU
f∘φf\circ\varphi is measurable on VV
Jφ(f∘φ)J_\varphi(f\circ\varphi) is bounded on VV
∣Jφ∣(f∘φ)|J_\varphi|(f\circ\varphi) is integrable on VV

$$|J_\varphi|(f\circ\varphi)$$ is integrable on $$V$$

Explanation

The integrability equivalence requires the absolute-Jacobian-weighted composition to be integrable on VV. Measurability alone, or boundedness of the signed-Jacobian product, is not the stated criterion.

21. Suppose μ\mu and ε\varepsilon are sigma-finite and f:E1×E2→[0,+∞]f:E_1\times E_2\to[0,+\infty] is measurable. What can be concluded about the function x↦∫E2f(x,y) ε(dy)x\mapsto\int_{E_2}f(x,y)\,\varepsilon(dy)?

It is continuous on E1E_1
It is measurable on E1E_1
It is integrable with respect to μ\mu
It is finite at every point of E1E_1

It is measurable on $$E_1$$

Explanation

Fubini-Tonelli ensures that integrating a nonnegative measurable function over the second coordinate produces a measurable function of the first coordinate. The result need not be finite, continuous, or integrable.

22. For sigma-finite measures μ,ε\mu,\varepsilon and measurable f:E1×E2→[0,+∞]f:E_1\times E_2\to[0,+\infty], which expression correctly computes the product-space integral by integrating first over E2E_2?

∫E2(∫E1f(x,y) ε(dy))μ(dx)\int_{E_2}\left(\int_{E_1}f(x,y)\,\varepsilon(dy)\right)\mu(dx)
(∫E1f(x,y) μ(dx))(∫E2f(x,y) ε(dy))\left(\int_{E_1}f(x,y)\,\mu(dx)\right)\left(\int_{E_2}f(x,y)\,\varepsilon(dy)\right)
∫E1(∫E2f(x,y) ε(dy))μ(dx)\int_{E_1}\left(\int_{E_2}f(x,y)\,\varepsilon(dy)\right)\mu(dx)
∫E2(∫E1f(x,y) μ(dx))ε(dy)\int_{E_2}\left(\int_{E_1}f(x,y)\,\mu(dx)\right)\varepsilon(dy)

$$\int_{E_1}\left(\int_{E_2}f(x,y)\,\varepsilon(dy)\right)\mu(dx)$$

Explanation

Integrating first over E2E_2 uses ε(dy)\varepsilon(dy) in the inner integral, followed by integration over E1E_1 with μ(dx)\mu(dx). The reversed iterated order is also valid, but it does not express the specified order; multiplying separate integrals is not the Fubini-Tonelli formula.

23. A function ff is integrable with respect to μ⊗ε\mu\otimes\varepsilon. For how many x∈E1x\in E_1 are its sections y↦f(x,y)y\mapsto f(x,y) guaranteed to be integrable with respect to ε\varepsilon?

For μ\mu-almost every xx
For every xx in E1E_1
For ε\varepsilon-almost every xx
For μ\mu-almost every yy

For $$\mu$$-almost every $$x$$

Explanation

Product-space integrability guarantees that the sections in the second variable are integrable for μ\mu-almost every first-coordinate value. This does not assert integrability at every point, and the exceptional set is measured by μ\mu, not ε\varepsilon.

24. Let independent random variables X and Y be given, with bounded measurable functions f and g. Which identity follows for their expectations?

E[f(X)g(Y)]=f(E[X])g(E[Y])\mathbb{E}[f(X)g(Y)] = f(\mathbb{E}[X])g(\mathbb{E}[Y])
E[f(X)g(Y)]=E[f(X)]+E[g(Y)]\mathbb{E}[f(X)g(Y)] = \mathbb{E}[f(X)] + \mathbb{E}[g(Y)]
E[f(X)g(Y)]=E[f(X)]E[g(Y)]\mathbb{E}[f(X)g(Y)] = \mathbb{E}[f(X)]\mathbb{E}[g(Y)]
E[f(X)g(Y)]=E[f(Y)]E[g(X)]\mathbb{E}[f(X)g(Y)] = \mathbb{E}[f(Y)]\mathbb{E}[g(X)]

$$\mathbb{E}[f(X)g(Y)] = \mathbb{E}[f(X)]\mathbb{E}[g(Y)]$$

Explanation

Independence makes the expectation of the product of bounded measurable functions factor into the product of their expectations. Adding expectations confuses product factorization with linearity, while the other expressions incorrectly swap or move function inputs.

25. For every pair of measurable sets A and B, a pair of random variables satisfies E[1A(X)1B(Y)]=E[1A(X)]E[1B(Y)]\mathbb{E}[\mathbf{1}_A(X)\mathbf{1}_B(Y)] = \mathbb{E}[\mathbf{1}_A(X)]\mathbb{E}[\mathbf{1}_B(Y)]. What does this establish?

X and Y have the same distribution
X and Y are independent only if the identity holds for every bounded real-valued function
X and Y are independent
X and Y are uncorrelated

X and Y are independent

Explanation

Indicator functions turn the expectation identity into the factorization of joint probabilities for all measurable sets, which establishes independence. Uncorrelatedness is a weaker moment condition, and identical distributions are not implied.

26. Which condition characterizes independence of X and Y using functions of the two variables?

Expectation factorizes for one pair of nonnegative measurable functions
Covariance is zero for every pair of bounded measurable functions
Expectation factorizes for every pair of nonnegative measurable functions, equivalently every pair of bounded measurable functions
Expectation factorization implies independence, but independence need not imply factorization

Expectation factorizes for every pair of nonnegative measurable functions, equivalently every pair of bounded measurable functions

Explanation

Independence is equivalent to expectation factorization for every pair of nonnegative measurable functions, or equivalently every pair of bounded measurable functions. Zero covariance is weaker, and testing just one pair of functions does not characterize independence.

27. For a random vector X in Rd\mathbb{R}^d, which expression gives its characteristic function at t∈Rdt \in \mathbb{R}^d?

φX(t)=E[ei⟨t,X⟩]=∫Rdei⟨t,x⟩PX(dx)\varphi_X(t)=\mathbb{E}[e^{i\langle t,X\rangle}]=\int_{\mathbb{R}^d}e^{i\langle t,x\rangle}P_X(dx)
φX(t)=E[e⟨t,X⟩]=∫Rde⟨t,x⟩PX(dx)\varphi_X(t)=\mathbb{E}[e^{\langle t,X\rangle}]=\int_{\mathbb{R}^d}e^{\langle t,x\rangle}P_X(dx)
φX(t)=ei⟨t,E[X]⟩\varphi_X(t)=e^{i\langle t,\mathbb{E}[X]\rangle}
φX(t)=E[ei∑k=1dtkXk2]\varphi_X(t)=\mathbb{E}[e^{i\sum_{k=1}^d t_kX_k^2}]

$$\varphi_X(t)=\mathbb{E}[e^{i\langle t,X\rangle}]=\int_{\mathbb{R}^d}e^{i\langle t,x\rangle}P_X(dx)$$

Explanation

The characteristic function is the expectation of the complex exponential of the inner product, equivalently its integral against the distribution of X. The real exponential is the moment-generating form, while the other expressions alter or omit the required averaging.

28. A random vector X=(X1,…,Xd)X=(X_1,\ldots,X_d) has characteristic function satisfying φX(t)=∏k=1dφXk(tk)\varphi_X(t)=\prod_{k=1}^d\varphi_{X_k}(t_k) for every t∈Rdt\in\mathbb{R}^d. What follows?

The components of X are independent
The components are independent only at the particular vector t tested
The components of X are identically distributed
The components of X are dependent

The components of X are independent

Explanation

The characteristic-function factorization for every vector t is equivalent to independence of the components. Factorization at just one t is insufficient, and independence does not require identical distributions.

29. For a pair of real-valued random variables, suppose their joint CDF satisfies F(X,Y)(x,y)=FX(x)FY(y)F_{(X,Y)}(x,y)=F_X(x)F_Y(y) for every real pair (x,y)(x,y). What follows?

X and Y are independent only if both have densities
X and Y are uncorrelated but not necessarily independent
X and Y have identical distributions
X and Y are independent

X and Y are independent

Explanation

Factorization of the joint CDF at every real pair is necessary and sufficient for independence. A density is not required, checking one pair is insufficient, and uncorrelatedness alone does not establish independence.

30. A pair (X,Y)(X,Y) has a joint density, and its joint density equals the product of its marginal densities for Lebesgue-almost every point in R2\mathbb{R}^2. What conclusion is warranted?

X and Y are independent
X and Y are independent only if the equality holds at every point
X and Y are independent only if their marginal CDFs also have densities
X and Y may be dependent because the equality is only almost everywhere

X and Y are independent

Explanation

For a joint law with a density, factorization into the marginal densities almost everywhere is equivalent to independence. Requiring equality everywhere is stronger than needed, marginal CDFs need not have densities, and the criterion is sufficient as well as necessary.

Study with flashcards

Lock in the answers with 65 flashcards on Product Integration and Change of Variables.

What generates the product σ-algebra E1⊗E2E_1 \otimes E_2 on E1×E2E_1 \times E_2?

Measurable rectangles A×BA \times B with A∈E1A \in E_1 and B∈E2B \in E_2.

If C∈E1⊗E2C \in E_1 \otimes E_2, where does each x-section CxC_x belong?

E2E_2.

If C∈E1⊗E2C \in E_1 \otimes E_2, where does each y-section CyC^y belong?

E1E_1.

See the flashcards →

Read the study guide

Go over the full study guide for Product Integration and Change of Variables.

Read the study guide →

Related classes

Make your own practice quizzes

Upload your class notes and Revizly writes a practice quiz with explained answers in about a minute.

Quiz maker