1244 CHAPTER 35. WEAK DERIVATIVES

35.6 Change Of Variables Formula Lipschitz MapsWith Rademacher’s theorem, one can give a general change of variables formula involvingLipschitz maps. First here is an elementary estimate.

Lemma 35.6.1 Suppose V is an n− 1 dimensional subspace of Rn and K is a compactsubset of V . Then letting

Kε ≡ ∪x∈KB(x,ε) = K +B(0,ε) ,

it follows thatmn (Kε)≤ 2n

ε (diam(K)+ ε)n−1 .

Proof: Let an orthonormal basis for V be {v1, · · · ,vn−1} and let

{v1, · · · ,vn−1,vn}

be an orthonormal basis for Rn. Now define a linear transformation, Q by Qvi = ei. ThusQQ∗ = Q∗Q = I and Q preserves all distances because∣∣∣∣∣Q∑

iaiei

∣∣∣∣∣2

=

∣∣∣∣∣∑iaivi

∣∣∣∣∣2

= ∑i|ai|2 =

∣∣∣∣∣∑iaiei

∣∣∣∣∣2

.

Letting k0 ∈ K, it follows K ⊆ B(k0,diam(K)) and so,

QK ⊆ Bn−1 (Qk0,diam(QK)) = Bn−1 (Qk0,diam(K))

where Bn−1 refers to the ball taken with respect to the usual norm in Rn−1. Every point ofKε is within ε of some point of K and so it follows that every point of QKε is within ε ofsome point of QK. Therefore,

QKε ⊆ Bn−1 (Qk0,diam(QK)+ ε)× (−ε,ε) ,

To see this, let x ∈ QKε . Then there exists k ∈ QK such that |k−x| < ε . Therefore,|(x1, · · · ,xn−1)− (k1, · · · ,kn−1)|< ε and |xn− kn|< ε and so x is contained in the set on theright in the above inclusion because kn = 0. However, the measure of the set on the right issmaller than

[2(diam(QK)+ ε)]n−1 (2ε) = 2n [(diam(K)+ ε)]n−1ε.

This proves the lemma.Next is the definition of a point of density. This is sort of like an interior point but not

as good.

Definition 35.6.2 Let E be a Lebesgue measurable set. x ∈ E is a point of density if

limr→0

m(E ∩B(x,r))m(B(x,r))

= 1.

1244 CHAPTER 35. WEAK DERIVATIVES35.6 Change Of Variables Formula Lipschitz MapsWith Rademacher’s theorem, one can give a general change of variables formula involvingLipschitz maps. First here is an elementary estimate.Lemma 35.6.1 Suppose V is an n—1 dimensional subspace of R” and K is a compactsubset of V. Then lettingKe = UxexB (x,€) = K +B (0,8),it follows thatMn (Ke) < 2"e (diam(K) +e)".Proof: Let an orthonormal basis for V be {v1 ,--- ,V,—1} and let{Vis »Vn—1, Vn}be an orthonormal basis for R”. Now define a linear transformation, Q by Qv; = e;. ThusQO* = Q*Q =/ and Q preserves all distances because2=) |a;’ =i2 2| OY ae; Yai Yi aie;i i iLetting ko € K, it follows K C B(ko,diam(K)) and so,OK CB"! (Oko, diam (QK)) = B"' (ko, diam (K))where B”~! refers to the ball taken with respect to the usual norm in R’~!. Every point ofKe is within € of some point of K and so it follows that every point of QK¢ is within € ofsome point of OK. Therefore,OK, © B" | (Qko,diam(QK) + €) x (—€,€),To see this, let x € QKeg. Then there exists k € QK such that |k—x| < €. Therefore,\(x1,°+* »Xn—1) — (A1,-++ Kn—1)| < € and |x, — k,| < € and so x is contained in the set on theright in the above inclusion because k, = 0. However, the measure of the set on the right issmaller than[2 (diam (QK) + e)|"! (2€) = 2” [(diam (K) + e)|"! E.This proves the lemma.Next is the definition of a point of density. This is sort of like an interior point but notas good.Definition 35.6.2 Let E be a Lebesgue measurable set. x € E is a point of density ifkim m(E O B(x,r)) 4r>0 m(B(x,r))