Real Analysis, Problem 3.2.14 The Radon Nikodym Theorem The Next CEO of Stack OverflowRadon–Nikodym derivative and “normal” derivativeCalculating Radon Nikodym derivativeRadon–Nikodym theorem: special caseA counter-example to Radon-Nikodym Theorem?Absolute continuity and Radon-Nikodym derivativeIs $sigma$-finiteness unnecessary for Radon Nikodym theorem?Lebesgue–Radon–Nikodym Theorem ExplanationReal Analysis, Folland Corollary 3.6, The Lebesgue-Radon-Nikodym TheoremWeakening the “positive $mu$” condition in Radon-Nikodym theoremCorollary of Radon-Nikodym theorem

What is the difference between 'contrib' and 'non-free' packages repositories?

Is it OK to decorate a log book cover?

Are British MPs missing the point, with these 'Indicative Votes'?

Is it correct to say moon starry nights?

How does a dynamic QR code work?

MT "will strike" & LXX "will watch carefully" (Gen 3:15)?

Is it okay to majorly distort historical facts while writing a fiction story?

How can the PCs determine if an item is a phylactery?

Why did early computer designers eschew integers?

Is a distribution that is normal, but highly skewed, considered Gaussian?

How to show a landlord what we have in savings?

Incomplete cube

My ex-girlfriend uses my Apple ID to login to her iPad, do I have to give her my Apple ID password to reset it?

What happens if you break a law in another country outside of that country?

What did the word "leisure" mean in late 18th Century usage?

Is a linearly independent set whose span is dense a Schauder basis?

Does Germany produce more waste than the US?

Which acid/base does a strong base/acid react when added to a buffer solution?

Early programmable calculators with RS-232

Can I cast Thunderwave and be at the center of its bottom face, but not be affected by it?

Direct Implications Between USA and UK in Event of No-Deal Brexit

pgfplots: How to draw a tangent graph below two others?

Is this a new Fibonacci Identity?

Creating a script with console commands



Real Analysis, Problem 3.2.14 The Radon Nikodym Theorem



The Next CEO of Stack OverflowRadon–Nikodym derivative and “normal” derivativeCalculating Radon Nikodym derivativeRadon–Nikodym theorem: special caseA counter-example to Radon-Nikodym Theorem?Absolute continuity and Radon-Nikodym derivativeIs $sigma$-finiteness unnecessary for Radon Nikodym theorem?Lebesgue–Radon–Nikodym Theorem ExplanationReal Analysis, Folland Corollary 3.6, The Lebesgue-Radon-Nikodym TheoremWeakening the “positive $mu$” condition in Radon-Nikodym theoremCorollary of Radon-Nikodym theorem










2












$begingroup$



Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Attempted proof - Consider $mu(E)$ where $E$ is a $sigma$-finite set for $nu$. Since we have that $mu$ is finite then clearly $mu(E)$ must be bounded which implies it has a supremum. We will define the supremum as $$L = supF: F textis sigma-textfinite for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since it is taken to be a countable union of $sigma$-finite sets. On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$. Now we apply the Radon-Nikodym theorem: so, suppose that $Fcap E = emptyset$ well since $nullmu$ then by definition $nu(F) = mu(F) = 0$. If $mu(F) > 0$ then for sake of contradiction suppose $|nu(F)|neq infty$. Then since $mu$ is finite $mu(Fcup E) > mu(E)$ then this implies that $Ecup F$ is not $sigma$-finite since a $sigma$-finite set with a union of a finite set is $sigma$-finite. Therefore $|nu(F)| = infty$.



Thus I believe we can refer to Radon-Nikidym theorem to conclude that there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$.



I am not sure if this is completely correct, any suggestions is greatly appreciated.










share|cite|improve this question









$endgroup$







  • 1




    $begingroup$
    If you're still not able to decide on your own whether your analysis proofs are correct, even after asking literally dozens of these proof-verification questions, you have much deeper issues that need to be resolved with your faculty / tutor/ mentor. Not a website.
    $endgroup$
    – T. Bongers
    Jul 15 '16 at 23:10















2












$begingroup$



Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Attempted proof - Consider $mu(E)$ where $E$ is a $sigma$-finite set for $nu$. Since we have that $mu$ is finite then clearly $mu(E)$ must be bounded which implies it has a supremum. We will define the supremum as $$L = supF: F textis sigma-textfinite for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since it is taken to be a countable union of $sigma$-finite sets. On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$. Now we apply the Radon-Nikodym theorem: so, suppose that $Fcap E = emptyset$ well since $nullmu$ then by definition $nu(F) = mu(F) = 0$. If $mu(F) > 0$ then for sake of contradiction suppose $|nu(F)|neq infty$. Then since $mu$ is finite $mu(Fcup E) > mu(E)$ then this implies that $Ecup F$ is not $sigma$-finite since a $sigma$-finite set with a union of a finite set is $sigma$-finite. Therefore $|nu(F)| = infty$.



Thus I believe we can refer to Radon-Nikidym theorem to conclude that there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$.



I am not sure if this is completely correct, any suggestions is greatly appreciated.










share|cite|improve this question









$endgroup$







  • 1




    $begingroup$
    If you're still not able to decide on your own whether your analysis proofs are correct, even after asking literally dozens of these proof-verification questions, you have much deeper issues that need to be resolved with your faculty / tutor/ mentor. Not a website.
    $endgroup$
    – T. Bongers
    Jul 15 '16 at 23:10













2












2








2


3



$begingroup$



Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Attempted proof - Consider $mu(E)$ where $E$ is a $sigma$-finite set for $nu$. Since we have that $mu$ is finite then clearly $mu(E)$ must be bounded which implies it has a supremum. We will define the supremum as $$L = supF: F textis sigma-textfinite for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since it is taken to be a countable union of $sigma$-finite sets. On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$. Now we apply the Radon-Nikodym theorem: so, suppose that $Fcap E = emptyset$ well since $nullmu$ then by definition $nu(F) = mu(F) = 0$. If $mu(F) > 0$ then for sake of contradiction suppose $|nu(F)|neq infty$. Then since $mu$ is finite $mu(Fcup E) > mu(E)$ then this implies that $Ecup F$ is not $sigma$-finite since a $sigma$-finite set with a union of a finite set is $sigma$-finite. Therefore $|nu(F)| = infty$.



Thus I believe we can refer to Radon-Nikidym theorem to conclude that there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$.



I am not sure if this is completely correct, any suggestions is greatly appreciated.










share|cite|improve this question









$endgroup$





Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Attempted proof - Consider $mu(E)$ where $E$ is a $sigma$-finite set for $nu$. Since we have that $mu$ is finite then clearly $mu(E)$ must be bounded which implies it has a supremum. We will define the supremum as $$L = supF: F textis sigma-textfinite for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since it is taken to be a countable union of $sigma$-finite sets. On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$. Now we apply the Radon-Nikodym theorem: so, suppose that $Fcap E = emptyset$ well since $nullmu$ then by definition $nu(F) = mu(F) = 0$. If $mu(F) > 0$ then for sake of contradiction suppose $|nu(F)|neq infty$. Then since $mu$ is finite $mu(Fcup E) > mu(E)$ then this implies that $Ecup F$ is not $sigma$-finite since a $sigma$-finite set with a union of a finite set is $sigma$-finite. Therefore $|nu(F)| = infty$.



Thus I believe we can refer to Radon-Nikidym theorem to conclude that there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$.



I am not sure if this is completely correct, any suggestions is greatly appreciated.







real-analysis measure-theory proof-verification






share|cite|improve this question













share|cite|improve this question











share|cite|improve this question




share|cite|improve this question










asked Jul 15 '16 at 22:59









WolfyWolfy

2,36111241




2,36111241







  • 1




    $begingroup$
    If you're still not able to decide on your own whether your analysis proofs are correct, even after asking literally dozens of these proof-verification questions, you have much deeper issues that need to be resolved with your faculty / tutor/ mentor. Not a website.
    $endgroup$
    – T. Bongers
    Jul 15 '16 at 23:10












  • 1




    $begingroup$
    If you're still not able to decide on your own whether your analysis proofs are correct, even after asking literally dozens of these proof-verification questions, you have much deeper issues that need to be resolved with your faculty / tutor/ mentor. Not a website.
    $endgroup$
    – T. Bongers
    Jul 15 '16 at 23:10







1




1




$begingroup$
If you're still not able to decide on your own whether your analysis proofs are correct, even after asking literally dozens of these proof-verification questions, you have much deeper issues that need to be resolved with your faculty / tutor/ mentor. Not a website.
$endgroup$
– T. Bongers
Jul 15 '16 at 23:10




$begingroup$
If you're still not able to decide on your own whether your analysis proofs are correct, even after asking literally dozens of these proof-verification questions, you have much deeper issues that need to be resolved with your faculty / tutor/ mentor. Not a website.
$endgroup$
– T. Bongers
Jul 15 '16 at 23:10










1 Answer
1






active

oldest

votes


















2












$begingroup$

@Wolfy, here is a detailed proof, following Folland's hints.




Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Proof - Let us work on hint a.).



Since $mu$ is a $sigma$-finite measure on $(X,M)$, there is a countable family $X_n_n$ of disjoint measurable subsets of $X$ such that, fo all $n$, $mu(X_n)<infty$ and $X=bigcup_n X_n$.



Now, let $P$ and $N$ be a Hahn decomposition of $X$ for $nu$. Then $X=Pcup N$, $Pcap N= emptyset$ and, for all measurable set $A$,
$$nu^+(A)=nu(Acap P) quad nu^-(A)=-nu(Acap N)$$



For each $n$, $nu^+|_X_nll mu|_X_n$ and $nu^-|_X_n ll mu|_X_n$.
If we can prove for those cases, that there are extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$.



It is easy to see that $g_n = 0$ $mu$-a.e. on $Pcap X_n$ and $f_n = 0$ $mu$-a.e. on $Ncap X_n$. So, $h_n=f_n-g_n$ is well defined $mu$-a.e.. (I means we will not have $infty -infty$).



Since $X_n_n$ is a countable family of disjoint measurable subsets of $X$, we have that
$h= sum_n h_n$ is a measurable function and $h:Xrightarrow [-infty,infty]$ and we also have, for any $A$ measurable set,
beginalign*
nu(A) &= sum_nnu(Acap X_ncap P) + sum_nnu(Acap X_ncap N)= \
&= sum_nnu^+(Acap X_n) - sum_nnu^-(Acap X_n)= \
&= sum_n int_Acap X_n f_n dmu - sum_n int_Acap X_n g_n dmu= \
&= int_A sum_n f_n dmu - int_A sum_n g_n dmu= \
&= int_A sum_n (f_n - g_n) dmu = \
& =int_A sum_n h_n dmu = int_A h dmu
endalign*

So we have proved that, for any $A$ measurable set,
$$nu(A) = int_A h dmu$$
Since $nu$ is a signed measure, it does not attain either $+infty$ or $-infty$. It follows then that $h$ is not only measurable but it an extended $mu$-integrable function. (In other word, either $int h^+ dmu <infty$ or $int h^- dmu <infty$ or both).



Now, to complete the proof, all we need is to prove that




If $nu$ is a (positive) measure and $mu$ is a finite measure on $(X,M)$ such that $null mu$, there exists an extended measurable function $f:Xrightarrow [0,infty]$ such that $dnu = fdmu$.




Consider the set
$$S = mu(F) : textrm $F$ is a $sigma$-finite set for $nu$ $$



Since we have that $mu$ is finite then clearly $S$ must be bounded which implies it has a finite supremum. We will define the supremum as
$$L = supmu(F) : F textis a sigmatext-finite set for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since $E$ is a $sigma$-finite set (being a countable union of $sigma$-finite sets). On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.
So we used hint b.). Now let us work on hint c.).



Now, since $nu|_E$ is $sigma$-finite and $nu|_Ell mu|_E$ we apply the Radon-Nikodym theorem, and so there is $f_1:Erightarrow [0,+infty)$ such that
$$nu(Acap E) =int_Acap E f_1 dmu tag1$$



Now, given any measurable set $F$ such that $Fcap E = emptyset$, suppose that $mu(F)>0$ and $nu(F) <infty$. Then $Ecup F$ is $sigma$-finite for $nu$, (that is $Ecup F in S$) and
$$mu(Ecup F) =mu(E) +mu( F) > mu(E)=L = sup S$$
Contradiction.



So, for any measurable set $F$ such that $Fcap E = emptyset$,
we have two cases:




  1. $mu(F) =0$ then, since $nu ll mu$, $nu(F)=0$.

  2. if $mu(F)>0$ then $nu(F)=+infty$.

So, for any measurable set $F$ such that $Fcap E = emptyset$,



$$ nu(F) = int_F +infty chi_E^c dmu $$



That means, for any measurable set $A$,



$$ nu(Acap E^c) = int_Acap E^c +infty chi_E^c dmu tag2$$



Let $f:X rightarrow [0,+infty]$ be defined as $f(x)=f_1(x)$ for $x in E$ and $f(x)=+infty$ for $x notin E$. From $(1)$ and $(2)$, we have, for any measurable set $A$,
$$nu(A)= nu(Acap E)+ nu(Acap E^c)= int_Acap E f_1 dmu +int_Acap E^c +infty chi_E^c dmu= int_A f dmu$$






share|cite|improve this answer











$endgroup$












  • $begingroup$
    Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
    $endgroup$
    – Wolfy
    Dec 29 '16 at 17:55










  • $begingroup$
    In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
    $endgroup$
    – user82261
    Mar 27 at 0:40











  • $begingroup$
    @user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
    $endgroup$
    – Ramiro
    Mar 27 at 11:26











  • $begingroup$
    @Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
    $endgroup$
    – user82261
    Mar 28 at 2:34










  • $begingroup$
    @user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
    $endgroup$
    – Ramiro
    Mar 28 at 10:26











Your Answer





StackExchange.ifUsing("editor", function ()
return StackExchange.using("mathjaxEditing", function ()
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix)
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
);
);
, "mathjax-editing");

StackExchange.ready(function()
var channelOptions =
tags: "".split(" "),
id: "69"
;
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function()
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled)
StackExchange.using("snippets", function()
createEditor();
);

else
createEditor();

);

function createEditor()
StackExchange.prepareEditor(
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader:
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
,
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
);



);













draft saved

draft discarded


















StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f1860803%2freal-analysis-problem-3-2-14-the-radon-nikodym-theorem%23new-answer', 'question_page');

);

Post as a guest















Required, but never shown

























1 Answer
1






active

oldest

votes








1 Answer
1






active

oldest

votes









active

oldest

votes






active

oldest

votes









2












$begingroup$

@Wolfy, here is a detailed proof, following Folland's hints.




Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Proof - Let us work on hint a.).



Since $mu$ is a $sigma$-finite measure on $(X,M)$, there is a countable family $X_n_n$ of disjoint measurable subsets of $X$ such that, fo all $n$, $mu(X_n)<infty$ and $X=bigcup_n X_n$.



Now, let $P$ and $N$ be a Hahn decomposition of $X$ for $nu$. Then $X=Pcup N$, $Pcap N= emptyset$ and, for all measurable set $A$,
$$nu^+(A)=nu(Acap P) quad nu^-(A)=-nu(Acap N)$$



For each $n$, $nu^+|_X_nll mu|_X_n$ and $nu^-|_X_n ll mu|_X_n$.
If we can prove for those cases, that there are extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$.



It is easy to see that $g_n = 0$ $mu$-a.e. on $Pcap X_n$ and $f_n = 0$ $mu$-a.e. on $Ncap X_n$. So, $h_n=f_n-g_n$ is well defined $mu$-a.e.. (I means we will not have $infty -infty$).



Since $X_n_n$ is a countable family of disjoint measurable subsets of $X$, we have that
$h= sum_n h_n$ is a measurable function and $h:Xrightarrow [-infty,infty]$ and we also have, for any $A$ measurable set,
beginalign*
nu(A) &= sum_nnu(Acap X_ncap P) + sum_nnu(Acap X_ncap N)= \
&= sum_nnu^+(Acap X_n) - sum_nnu^-(Acap X_n)= \
&= sum_n int_Acap X_n f_n dmu - sum_n int_Acap X_n g_n dmu= \
&= int_A sum_n f_n dmu - int_A sum_n g_n dmu= \
&= int_A sum_n (f_n - g_n) dmu = \
& =int_A sum_n h_n dmu = int_A h dmu
endalign*

So we have proved that, for any $A$ measurable set,
$$nu(A) = int_A h dmu$$
Since $nu$ is a signed measure, it does not attain either $+infty$ or $-infty$. It follows then that $h$ is not only measurable but it an extended $mu$-integrable function. (In other word, either $int h^+ dmu <infty$ or $int h^- dmu <infty$ or both).



Now, to complete the proof, all we need is to prove that




If $nu$ is a (positive) measure and $mu$ is a finite measure on $(X,M)$ such that $null mu$, there exists an extended measurable function $f:Xrightarrow [0,infty]$ such that $dnu = fdmu$.




Consider the set
$$S = mu(F) : textrm $F$ is a $sigma$-finite set for $nu$ $$



Since we have that $mu$ is finite then clearly $S$ must be bounded which implies it has a finite supremum. We will define the supremum as
$$L = supmu(F) : F textis a sigmatext-finite set for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since $E$ is a $sigma$-finite set (being a countable union of $sigma$-finite sets). On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.
So we used hint b.). Now let us work on hint c.).



Now, since $nu|_E$ is $sigma$-finite and $nu|_Ell mu|_E$ we apply the Radon-Nikodym theorem, and so there is $f_1:Erightarrow [0,+infty)$ such that
$$nu(Acap E) =int_Acap E f_1 dmu tag1$$



Now, given any measurable set $F$ such that $Fcap E = emptyset$, suppose that $mu(F)>0$ and $nu(F) <infty$. Then $Ecup F$ is $sigma$-finite for $nu$, (that is $Ecup F in S$) and
$$mu(Ecup F) =mu(E) +mu( F) > mu(E)=L = sup S$$
Contradiction.



So, for any measurable set $F$ such that $Fcap E = emptyset$,
we have two cases:




  1. $mu(F) =0$ then, since $nu ll mu$, $nu(F)=0$.

  2. if $mu(F)>0$ then $nu(F)=+infty$.

So, for any measurable set $F$ such that $Fcap E = emptyset$,



$$ nu(F) = int_F +infty chi_E^c dmu $$



That means, for any measurable set $A$,



$$ nu(Acap E^c) = int_Acap E^c +infty chi_E^c dmu tag2$$



Let $f:X rightarrow [0,+infty]$ be defined as $f(x)=f_1(x)$ for $x in E$ and $f(x)=+infty$ for $x notin E$. From $(1)$ and $(2)$, we have, for any measurable set $A$,
$$nu(A)= nu(Acap E)+ nu(Acap E^c)= int_Acap E f_1 dmu +int_Acap E^c +infty chi_E^c dmu= int_A f dmu$$






share|cite|improve this answer











$endgroup$












  • $begingroup$
    Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
    $endgroup$
    – Wolfy
    Dec 29 '16 at 17:55










  • $begingroup$
    In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
    $endgroup$
    – user82261
    Mar 27 at 0:40











  • $begingroup$
    @user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
    $endgroup$
    – Ramiro
    Mar 27 at 11:26











  • $begingroup$
    @Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
    $endgroup$
    – user82261
    Mar 28 at 2:34










  • $begingroup$
    @user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
    $endgroup$
    – Ramiro
    Mar 28 at 10:26















2












$begingroup$

@Wolfy, here is a detailed proof, following Folland's hints.




Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Proof - Let us work on hint a.).



Since $mu$ is a $sigma$-finite measure on $(X,M)$, there is a countable family $X_n_n$ of disjoint measurable subsets of $X$ such that, fo all $n$, $mu(X_n)<infty$ and $X=bigcup_n X_n$.



Now, let $P$ and $N$ be a Hahn decomposition of $X$ for $nu$. Then $X=Pcup N$, $Pcap N= emptyset$ and, for all measurable set $A$,
$$nu^+(A)=nu(Acap P) quad nu^-(A)=-nu(Acap N)$$



For each $n$, $nu^+|_X_nll mu|_X_n$ and $nu^-|_X_n ll mu|_X_n$.
If we can prove for those cases, that there are extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$.



It is easy to see that $g_n = 0$ $mu$-a.e. on $Pcap X_n$ and $f_n = 0$ $mu$-a.e. on $Ncap X_n$. So, $h_n=f_n-g_n$ is well defined $mu$-a.e.. (I means we will not have $infty -infty$).



Since $X_n_n$ is a countable family of disjoint measurable subsets of $X$, we have that
$h= sum_n h_n$ is a measurable function and $h:Xrightarrow [-infty,infty]$ and we also have, for any $A$ measurable set,
beginalign*
nu(A) &= sum_nnu(Acap X_ncap P) + sum_nnu(Acap X_ncap N)= \
&= sum_nnu^+(Acap X_n) - sum_nnu^-(Acap X_n)= \
&= sum_n int_Acap X_n f_n dmu - sum_n int_Acap X_n g_n dmu= \
&= int_A sum_n f_n dmu - int_A sum_n g_n dmu= \
&= int_A sum_n (f_n - g_n) dmu = \
& =int_A sum_n h_n dmu = int_A h dmu
endalign*

So we have proved that, for any $A$ measurable set,
$$nu(A) = int_A h dmu$$
Since $nu$ is a signed measure, it does not attain either $+infty$ or $-infty$. It follows then that $h$ is not only measurable but it an extended $mu$-integrable function. (In other word, either $int h^+ dmu <infty$ or $int h^- dmu <infty$ or both).



Now, to complete the proof, all we need is to prove that




If $nu$ is a (positive) measure and $mu$ is a finite measure on $(X,M)$ such that $null mu$, there exists an extended measurable function $f:Xrightarrow [0,infty]$ such that $dnu = fdmu$.




Consider the set
$$S = mu(F) : textrm $F$ is a $sigma$-finite set for $nu$ $$



Since we have that $mu$ is finite then clearly $S$ must be bounded which implies it has a finite supremum. We will define the supremum as
$$L = supmu(F) : F textis a sigmatext-finite set for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since $E$ is a $sigma$-finite set (being a countable union of $sigma$-finite sets). On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.
So we used hint b.). Now let us work on hint c.).



Now, since $nu|_E$ is $sigma$-finite and $nu|_Ell mu|_E$ we apply the Radon-Nikodym theorem, and so there is $f_1:Erightarrow [0,+infty)$ such that
$$nu(Acap E) =int_Acap E f_1 dmu tag1$$



Now, given any measurable set $F$ such that $Fcap E = emptyset$, suppose that $mu(F)>0$ and $nu(F) <infty$. Then $Ecup F$ is $sigma$-finite for $nu$, (that is $Ecup F in S$) and
$$mu(Ecup F) =mu(E) +mu( F) > mu(E)=L = sup S$$
Contradiction.



So, for any measurable set $F$ such that $Fcap E = emptyset$,
we have two cases:




  1. $mu(F) =0$ then, since $nu ll mu$, $nu(F)=0$.

  2. if $mu(F)>0$ then $nu(F)=+infty$.

So, for any measurable set $F$ such that $Fcap E = emptyset$,



$$ nu(F) = int_F +infty chi_E^c dmu $$



That means, for any measurable set $A$,



$$ nu(Acap E^c) = int_Acap E^c +infty chi_E^c dmu tag2$$



Let $f:X rightarrow [0,+infty]$ be defined as $f(x)=f_1(x)$ for $x in E$ and $f(x)=+infty$ for $x notin E$. From $(1)$ and $(2)$, we have, for any measurable set $A$,
$$nu(A)= nu(Acap E)+ nu(Acap E^c)= int_Acap E f_1 dmu +int_Acap E^c +infty chi_E^c dmu= int_A f dmu$$






share|cite|improve this answer











$endgroup$












  • $begingroup$
    Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
    $endgroup$
    – Wolfy
    Dec 29 '16 at 17:55










  • $begingroup$
    In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
    $endgroup$
    – user82261
    Mar 27 at 0:40











  • $begingroup$
    @user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
    $endgroup$
    – Ramiro
    Mar 27 at 11:26











  • $begingroup$
    @Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
    $endgroup$
    – user82261
    Mar 28 at 2:34










  • $begingroup$
    @user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
    $endgroup$
    – Ramiro
    Mar 28 at 10:26













2












2








2





$begingroup$

@Wolfy, here is a detailed proof, following Folland's hints.




Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Proof - Let us work on hint a.).



Since $mu$ is a $sigma$-finite measure on $(X,M)$, there is a countable family $X_n_n$ of disjoint measurable subsets of $X$ such that, fo all $n$, $mu(X_n)<infty$ and $X=bigcup_n X_n$.



Now, let $P$ and $N$ be a Hahn decomposition of $X$ for $nu$. Then $X=Pcup N$, $Pcap N= emptyset$ and, for all measurable set $A$,
$$nu^+(A)=nu(Acap P) quad nu^-(A)=-nu(Acap N)$$



For each $n$, $nu^+|_X_nll mu|_X_n$ and $nu^-|_X_n ll mu|_X_n$.
If we can prove for those cases, that there are extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$.



It is easy to see that $g_n = 0$ $mu$-a.e. on $Pcap X_n$ and $f_n = 0$ $mu$-a.e. on $Ncap X_n$. So, $h_n=f_n-g_n$ is well defined $mu$-a.e.. (I means we will not have $infty -infty$).



Since $X_n_n$ is a countable family of disjoint measurable subsets of $X$, we have that
$h= sum_n h_n$ is a measurable function and $h:Xrightarrow [-infty,infty]$ and we also have, for any $A$ measurable set,
beginalign*
nu(A) &= sum_nnu(Acap X_ncap P) + sum_nnu(Acap X_ncap N)= \
&= sum_nnu^+(Acap X_n) - sum_nnu^-(Acap X_n)= \
&= sum_n int_Acap X_n f_n dmu - sum_n int_Acap X_n g_n dmu= \
&= int_A sum_n f_n dmu - int_A sum_n g_n dmu= \
&= int_A sum_n (f_n - g_n) dmu = \
& =int_A sum_n h_n dmu = int_A h dmu
endalign*

So we have proved that, for any $A$ measurable set,
$$nu(A) = int_A h dmu$$
Since $nu$ is a signed measure, it does not attain either $+infty$ or $-infty$. It follows then that $h$ is not only measurable but it an extended $mu$-integrable function. (In other word, either $int h^+ dmu <infty$ or $int h^- dmu <infty$ or both).



Now, to complete the proof, all we need is to prove that




If $nu$ is a (positive) measure and $mu$ is a finite measure on $(X,M)$ such that $null mu$, there exists an extended measurable function $f:Xrightarrow [0,infty]$ such that $dnu = fdmu$.




Consider the set
$$S = mu(F) : textrm $F$ is a $sigma$-finite set for $nu$ $$



Since we have that $mu$ is finite then clearly $S$ must be bounded which implies it has a finite supremum. We will define the supremum as
$$L = supmu(F) : F textis a sigmatext-finite set for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since $E$ is a $sigma$-finite set (being a countable union of $sigma$-finite sets). On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.
So we used hint b.). Now let us work on hint c.).



Now, since $nu|_E$ is $sigma$-finite and $nu|_Ell mu|_E$ we apply the Radon-Nikodym theorem, and so there is $f_1:Erightarrow [0,+infty)$ such that
$$nu(Acap E) =int_Acap E f_1 dmu tag1$$



Now, given any measurable set $F$ such that $Fcap E = emptyset$, suppose that $mu(F)>0$ and $nu(F) <infty$. Then $Ecup F$ is $sigma$-finite for $nu$, (that is $Ecup F in S$) and
$$mu(Ecup F) =mu(E) +mu( F) > mu(E)=L = sup S$$
Contradiction.



So, for any measurable set $F$ such that $Fcap E = emptyset$,
we have two cases:




  1. $mu(F) =0$ then, since $nu ll mu$, $nu(F)=0$.

  2. if $mu(F)>0$ then $nu(F)=+infty$.

So, for any measurable set $F$ such that $Fcap E = emptyset$,



$$ nu(F) = int_F +infty chi_E^c dmu $$



That means, for any measurable set $A$,



$$ nu(Acap E^c) = int_Acap E^c +infty chi_E^c dmu tag2$$



Let $f:X rightarrow [0,+infty]$ be defined as $f(x)=f_1(x)$ for $x in E$ and $f(x)=+infty$ for $x notin E$. From $(1)$ and $(2)$, we have, for any measurable set $A$,
$$nu(A)= nu(Acap E)+ nu(Acap E^c)= int_Acap E f_1 dmu +int_Acap E^c +infty chi_E^c dmu= int_A f dmu$$






share|cite|improve this answer











$endgroup$



@Wolfy, here is a detailed proof, following Folland's hints.




Problem 3.3.14 - If $nu$ is an arbitrary signed measure and $mu$ is a $sigma$-finite measure on $(X,M)$ such that $null mu$, there exists an extended $mu$-integrable function $f:Xrightarrow [-infty,infty]$ such that $dnu = fdmu$. Hints:



a.) It suffices to assume that $mu$ is finite and $nu$ is positive.



b.) With these assumptions, there exists an $Ein M$ that is $sigma$-finite for $nu$ such that $mu(E)geq mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.



c.) The Radon-Nikodym theorem applies on $E$. If $Fcap E = emptyset$, then either $nu(F) = mu(F) = 0$ or $mu(F) > 0$ and $|nu(F)| = infty$.




Proof - Let us work on hint a.).



Since $mu$ is a $sigma$-finite measure on $(X,M)$, there is a countable family $X_n_n$ of disjoint measurable subsets of $X$ such that, fo all $n$, $mu(X_n)<infty$ and $X=bigcup_n X_n$.



Now, let $P$ and $N$ be a Hahn decomposition of $X$ for $nu$. Then $X=Pcup N$, $Pcap N= emptyset$ and, for all measurable set $A$,
$$nu^+(A)=nu(Acap P) quad nu^-(A)=-nu(Acap N)$$



For each $n$, $nu^+|_X_nll mu|_X_n$ and $nu^-|_X_n ll mu|_X_n$.
If we can prove for those cases, that there are extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$.



It is easy to see that $g_n = 0$ $mu$-a.e. on $Pcap X_n$ and $f_n = 0$ $mu$-a.e. on $Ncap X_n$. So, $h_n=f_n-g_n$ is well defined $mu$-a.e.. (I means we will not have $infty -infty$).



Since $X_n_n$ is a countable family of disjoint measurable subsets of $X$, we have that
$h= sum_n h_n$ is a measurable function and $h:Xrightarrow [-infty,infty]$ and we also have, for any $A$ measurable set,
beginalign*
nu(A) &= sum_nnu(Acap X_ncap P) + sum_nnu(Acap X_ncap N)= \
&= sum_nnu^+(Acap X_n) - sum_nnu^-(Acap X_n)= \
&= sum_n int_Acap X_n f_n dmu - sum_n int_Acap X_n g_n dmu= \
&= int_A sum_n f_n dmu - int_A sum_n g_n dmu= \
&= int_A sum_n (f_n - g_n) dmu = \
& =int_A sum_n h_n dmu = int_A h dmu
endalign*

So we have proved that, for any $A$ measurable set,
$$nu(A) = int_A h dmu$$
Since $nu$ is a signed measure, it does not attain either $+infty$ or $-infty$. It follows then that $h$ is not only measurable but it an extended $mu$-integrable function. (In other word, either $int h^+ dmu <infty$ or $int h^- dmu <infty$ or both).



Now, to complete the proof, all we need is to prove that




If $nu$ is a (positive) measure and $mu$ is a finite measure on $(X,M)$ such that $null mu$, there exists an extended measurable function $f:Xrightarrow [0,infty]$ such that $dnu = fdmu$.




Consider the set
$$S = mu(F) : textrm $F$ is a $sigma$-finite set for $nu$ $$



Since we have that $mu$ is finite then clearly $S$ must be bounded which implies it has a finite supremum. We will define the supremum as
$$L = supmu(F) : F textis a sigmatext-finite set for nu$$
Now lets take a sequence $E_n_1^infty$ that are $sigma$-finite with respect to $nu$ and $mu(E_n)rightarrow L$ as $nrightarrow infty$. Now let $$E = bigcup_1^inftyE_n$$ then $mu(E) leq L$ since $E$ is a $sigma$-finite set (being a countable union of $sigma$-finite sets). On the other hand, $mu(E)geq mu(E_n)$ for all $n$ and since $mu(E_n)rightarrow L$ then $mu(E) = L$. Now since $mu(E) = L$ then clearly by definition of $L$ we have that $mu(E) > mu(F)$ for all sets $F$ that are $sigma$-finite for $nu$.
So we used hint b.). Now let us work on hint c.).



Now, since $nu|_E$ is $sigma$-finite and $nu|_Ell mu|_E$ we apply the Radon-Nikodym theorem, and so there is $f_1:Erightarrow [0,+infty)$ such that
$$nu(Acap E) =int_Acap E f_1 dmu tag1$$



Now, given any measurable set $F$ such that $Fcap E = emptyset$, suppose that $mu(F)>0$ and $nu(F) <infty$. Then $Ecup F$ is $sigma$-finite for $nu$, (that is $Ecup F in S$) and
$$mu(Ecup F) =mu(E) +mu( F) > mu(E)=L = sup S$$
Contradiction.



So, for any measurable set $F$ such that $Fcap E = emptyset$,
we have two cases:




  1. $mu(F) =0$ then, since $nu ll mu$, $nu(F)=0$.

  2. if $mu(F)>0$ then $nu(F)=+infty$.

So, for any measurable set $F$ such that $Fcap E = emptyset$,



$$ nu(F) = int_F +infty chi_E^c dmu $$



That means, for any measurable set $A$,



$$ nu(Acap E^c) = int_Acap E^c +infty chi_E^c dmu tag2$$



Let $f:X rightarrow [0,+infty]$ be defined as $f(x)=f_1(x)$ for $x in E$ and $f(x)=+infty$ for $x notin E$. From $(1)$ and $(2)$, we have, for any measurable set $A$,
$$nu(A)= nu(Acap E)+ nu(Acap E^c)= int_Acap E f_1 dmu +int_Acap E^c +infty chi_E^c dmu= int_A f dmu$$







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Mar 28 at 10:21

























answered Jul 16 '16 at 3:29









RamiroRamiro

7,32421535




7,32421535











  • $begingroup$
    Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
    $endgroup$
    – Wolfy
    Dec 29 '16 at 17:55










  • $begingroup$
    In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
    $endgroup$
    – user82261
    Mar 27 at 0:40











  • $begingroup$
    @user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
    $endgroup$
    – Ramiro
    Mar 27 at 11:26











  • $begingroup$
    @Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
    $endgroup$
    – user82261
    Mar 28 at 2:34










  • $begingroup$
    @user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
    $endgroup$
    – Ramiro
    Mar 28 at 10:26
















  • $begingroup$
    Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
    $endgroup$
    – Wolfy
    Dec 29 '16 at 17:55










  • $begingroup$
    In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
    $endgroup$
    – user82261
    Mar 27 at 0:40











  • $begingroup$
    @user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
    $endgroup$
    – Ramiro
    Mar 27 at 11:26











  • $begingroup$
    @Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
    $endgroup$
    – user82261
    Mar 28 at 2:34










  • $begingroup$
    @user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
    $endgroup$
    – Ramiro
    Mar 28 at 10:26















$begingroup$
Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
$endgroup$
– Wolfy
Dec 29 '16 at 17:55




$begingroup$
Hey Ramiro, I am just curious about your intuition with considering the set $S$. I follow everything up until that point, then I get a bit lost.
$endgroup$
– Wolfy
Dec 29 '16 at 17:55












$begingroup$
In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
$endgroup$
– user82261
Mar 27 at 0:40





$begingroup$
In the expression for $nu(A)$ in part (a), how are you able to take $sum_n=1^infty$ inside the integral. One can arbitrarily change summations and integrals when the $f_n$'s are non-negative. But in this case, $f_n$ is real valued. By the Dominated Convergence Theorem, don't we need $sum_n=1^infty int_A |f_n| dmu < infty$?
$endgroup$
– user82261
Mar 27 at 0:40













$begingroup$
@user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
$endgroup$
– Ramiro
Mar 27 at 11:26





$begingroup$
@user82261 In part (a), we assume to have extended $mu$-integrable functions $f_n:Xrightarrow [0,infty]$ such that $dnu^+|_X_n = f_ndmu|_X_n$ and extended $mu$-integrable functions $g_n:Xrightarrow [0,infty]$ such that $dnu^-|_X_n = g_ndmu|_X_n$. So $f_n$ and $g_n$ are non-negative functions.
$endgroup$
– Ramiro
Mar 27 at 11:26













$begingroup$
@Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
$endgroup$
– user82261
Mar 28 at 2:34




$begingroup$
@Ramiro Right but $h_n$ is a real valued extended $mu-$ integrable function, right? You have used it in the expression where you changed the order of the sum and the integral? Or is that correct way to think about it is to first decompose $h_n$ into $f_n - g_n$, apply the result to each of these, and then combine them back again?
$endgroup$
– user82261
Mar 28 at 2:34












$begingroup$
@user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
$endgroup$
– Ramiro
Mar 28 at 10:26




$begingroup$
@user82261 We first decompose $h_n$. I have edited the steps to make it clearer. Let me know if you have any further question.
$endgroup$
– Ramiro
Mar 28 at 10:26

















draft saved

draft discarded
















































Thanks for contributing an answer to Mathematics Stack Exchange!


  • Please be sure to answer the question. Provide details and share your research!

But avoid


  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.

Use MathJax to format equations. MathJax reference.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f1860803%2freal-analysis-problem-3-2-14-the-radon-nikodym-theorem%23new-answer', 'question_page');

);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

Triangular numbers and gcdProving sum of a set is $0 pmod n$ if $n$ is odd, or $fracn2 pmod n$ if $n$ is even?Is greatest common divisor of two numbers really their smallest linear combination?GCD, LCM RelationshipProve a set of nonnegative integers with greatest common divisor 1 and closed under addition has all but finite many nonnegative integers.all pairs of a and b in an equation containing gcdTriangular Numbers Modulo $k$ - Hit All Values?Understanding the Existence and Uniqueness of the GCDGCD and LCM with logical symbolsThe greatest common divisor of two positive integers less than 100 is equal to 3. Their least common multiple is twelve times one of the integers.Suppose that for all integers $x$, $x|a$ and $x|b$ if and only if $x|c$. Then $c = gcd(a,b)$Which is the gcd of 2 numbers which are multiplied and the result is 600000?

Србија Садржај Етимологија Географија Историја Политички систем и уставно-правно уређење Становништво Привреда Образовање Култура Спорт Државни празници Галерија Напомене Референце Литература Спољашње везе Мени за навигацију44°48′N 20°28′E / 44.800° СГШ; 20.467° ИГД / 44.800; 20.46744°48′N 20°28′E / 44.800° СГШ; 20.467° ИГД / 44.800; 20.467ууРезултати пописа 2011. према старости и полуу„Положај, рељеф и клима”„Europe: Serbia”„Основни подаци”„Gross domestic product based on purchasing-power-parity (PPP) valuation of country GDP”„Human Development Report 2018 – "Human Development Indices and Indicators 6”„Устав Републике Србије”Правопис српскога језикаGoogle DriveComparative Hungarian Cultural StudiesCalcium and Magnesium in Groundwater: Occurrence and Significance for Human Health„UNSD — Methodology”„Процене становништва | Републички завод за статистику Србије”The Age of Nepotism: Travel Journals and Observations from the Balkans During the Depression„The Serbian Revolution and the Serbian State”„Устав Србије”„Serbia a few steps away from concluding WTO accession negotiations”„A credible enlargement perspective for and enhanced EU engagement with the Western Balkans”„Freedom in the World 2017”„Serbia: On the Way to EU Accession”„Human Development Indices and Indicators: 2018 Statistical Update”„2018 Social Progress Index”„Global Peace Index”Sabres of Two Easts: An Untold History of Muslims in Eastern Europe, Their Friends and Foes„Пројекат Растко—Лузица”„Serbia: Introduction”„Serbia”оригинала„The World Factbook: Serbia”„The World Factbook: Kosovo”„Border Police Department”„Uredba o kontroli prelaska administrativne linije prema Autonomnoj pokrajini Kosovo i Metohija”оригиналаIvana Carevic, Velimir Jovanovic, STRATIGRAPHIC-STRUCTURAL CHARACTERISTICS OF MAČVA BASIN, UDC 911.2:551.7(497.11), pp. 1Archived„About the Carpathians – Carpathian Heritage Society”оригинала„O Srbiji”оригинала„Статистички годишњак Србије, 2009: Географски прегледГеографија за осми разред основне школе„Отворена, електронска база едукационих радова”„Влада Републике Србије: Положај, рељеф и клима”„Копрен (Стара планина)”„Туристичка дестинација-Србија”„Висина водопада”„РХМЗ — Републички Хидрометеоролошки завод Србије Кнеза Вишеслава 66 Београд”„Фауна Србије”„Српске шуме на издисају”„Lepih šest odsto Srbije”„Илустрована историја Срба — Увод”„Винчанска култура - Градска општина Гроцка”„''„Винча — Праисторијска метропола”''”оригиналаЈужни Словени под византијском влашћу (600—1025)Држава маћедонских Словена„Карађорђе истина и мит, Проф. др Радош Љушић, Вечерње новости, фељтон, 18 наставака, 24. август - 10. септембар 2003.”„Политика: Како је утврђена војна неутралност, 13. јануар. 2010, приступљено децембра 2012.”„Србија и РС оживеле Дејтонски споразум”„Са српским пасошем у 104 земље”Војска Србије | О Војсци | Војска Србије — Улога, намена и задациАрхивираноВојска Србије | ОрганизацијаАрхивираноОдлука о изради Стратегије просторног развоја Републике Србије до 2020. годинеЗакон о територијалној организацији Републике СрбијеЗакон о државној управиНајчешће постављана питања.„Смањење броја статистичких региона кроз измене Закона о регионалном развоју”„2011 Human development Report”„Službena upotreba jezika i pisama”„Попис становништва, домаћинстава и станова 2011. године у Републици Србији. Књига 4: Вероисповест, матерњи језик и национална припадност”„Вероисповест, матерњи језик и национална”„Специјална известитељка УН за слободу религије и вероисповести Асма Јахангир, код Заштитника грађана Саше Јанковића”„Закон о државним и другим празницима у Републици Србији”„Веронаука у српским школама”„Serbia – Ancestral Genography Atlas”Бела књига Милошевићеве владавинеоригиналаGross domestic product based on purchasing-power-parity (PPP) per capita GDP БДП 2007—2013Актуелни показатељи — Република Србија„Попис становништва, домаћинстава и станова 2011. године у Републици Србији Књига 7: Економска активност”Zemlje kandidati za članstvo u EU„Putin drops South Stream gas pipeline to EU, courts Turkey”„„Соко — историјат””оригинала„„Рембас — историјат””оригинала„„Лубница — историјат””оригинала„„Штаваљ — Историјат””оригинала„„Боговина — историјат””оригинала„„Јасеновац — историјат””оригинала„„Вршка чука — историјат””оригинала„„Ибарски рудници — историјат””оригинала„Закон о просторном плану Републике Србије од 2010 до 2020”„Кривични законик — Недозвољена изградња нуклеарних постројења, члан 267”„Б92: Srbija uklonila obogaćeni uranijum, 25. октобар 2011”„Коришћење енергије ветра у Србији — природни услови и практична примена”„Енергија ветра”„Србија може да прави струју од сунца, биомасе, воде и ветра”„Моја електрана и друге ветрењаче”„Биомаса, струја без инвестиција”„Auto-karte Srbije”„www.srbija.gov.rs Статистике о Србији”оригинала„Статистика зе месец децембар и 2016. годину”„Turizam u Srbiji”„Univerzitet u Beogradu: Vek i po akademskog znanja”„Vojnomedicinska akademija: 165 godina tradicije i napretka”Никола Гиљен, Соња Јовићевић Јов и Јелена Мандић: Мирослављево јеванђеље; Текст је публикован у ревији „Историја” и настао је као део научно-истраживачког рада Фонда „Принцеза Оливера”„World music асоцијација Србије”оригинала„World music у Србији”оригинала„Pogledajte: Boban Marković svira u redakciji „Blica”!”„Eurovision Song Contest 2007 Final”„Projekat Rastko, Alojz Ujes: Joakim Vujic”„Унеско”„Списак локалитета Светске баштине”„Guča i Egzit zaludeli svet”оригинала„Sabor trubača GUČA”„Interesting facts about Exit”оригинала„FIFA Association Information”„Serbia women win EuroBasket title, gain first Olympics berth”„Odbojkašice ispisale istoriju – Srbija je svetski prvak!”„Сајт Ватерполо савеза Србије, Освојене медаље”„Сајт ФК Црвена звезда, Бари”„Сајт ФК Црвена звезда, Токио”„Blic:Zlatna Milica! Mandićeva donela Srbiji najsjajnije odličje u Londonu!”„Милица Мандић освојила златну медаљу („Политика”, 12. август 2012)”„Златни Давор Штефанек”„DŽUDO ŠAMPIONAT Majdov osvojio svetsko zlato”„Španovićeva trećim skokom svih vremena do zlata!”„Чудо Иване Шпановић — 7,24 м („Политика”, 5. март 2017)”The Age of Nepotism: Travel Journals and Observations from the Balkans During the DepressionCalcium and Magnesium in Groundwater: Occurrence and Significance for Human HealthComparative Hungarian Cultural StudiesБела књига Милошевићеве владавинеоригиналаComparative Hungarian Cultural StudiesSabres of Two Easts: An Untold History of Muslims in Eastern Europe, Their Friends and FoesГеографија за осми разред основне школеSerbia: the country, people, life, customsМедијиВодичПодациВлада Републике СрбијеНародна скупштина Републике СрбијеНародна канцеларија председника Републике СрбијеНародна банка СрбијеТуристичка организација СрбијеПортал еУправе Републике СрбијеРепубличко јавно правобранилаштвоууууууWorldCat151202876n851959190000 0000 9526 67094054598-24101000570825ge130919

Barbados Ynhâld Skiednis | Geografy | Demografy | Navigaasjemenu