Total variation distance of probaiblity measures
$begingroup$
Let $E$ be a set and $mathcal Esubseteq 2^E$ with $emptysetinmathcal E$. If $mu:mathcal Etomathbb R$ with $mu(emptyset)=0$, then $$operatorname{Var}_mu(B):=supleft{sum_{i=1}^nleft|mu(B_i)right|:ninmathbb Ntext{ and }B_1,ldots,B_ninmathcal Btext{ are disjoint with }biguplus_{i=1}^nB_isubseteq Bright}$$ for $Bsubseteq E$. If $Einmathcal E$, let $$left|muright|:=operatorname{Var}_mu(E).$$
Assume $(E,mathcal E)$ is a measurable space and $mu,nu$ are probability measures on $(E,mathcal E)$. Is it possible to show that $$left|mu-nuright|=sup_{Binmathcal E}left|mu(B)-nu(B)right|tag1$$ or is there a counterexample?
I was able to show the claim assuming that $mu$ and $nu$ are both absolutely continuous with respect to a common reference measure. Is it possible to show $(1)$ in general?
probability-theory measure-theory total-variation
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
$endgroup$
add a comment |
$begingroup$
Let $E$ be a set and $mathcal Esubseteq 2^E$ with $emptysetinmathcal E$. If $mu:mathcal Etomathbb R$ with $mu(emptyset)=0$, then $$operatorname{Var}_mu(B):=supleft{sum_{i=1}^nleft|mu(B_i)right|:ninmathbb Ntext{ and }B_1,ldots,B_ninmathcal Btext{ are disjoint with }biguplus_{i=1}^nB_isubseteq Bright}$$ for $Bsubseteq E$. If $Einmathcal E$, let $$left|muright|:=operatorname{Var}_mu(E).$$
Assume $(E,mathcal E)$ is a measurable space and $mu,nu$ are probability measures on $(E,mathcal E)$. Is it possible to show that $$left|mu-nuright|=sup_{Binmathcal E}left|mu(B)-nu(B)right|tag1$$ or is there a counterexample?
I was able to show the claim assuming that $mu$ and $nu$ are both absolutely continuous with respect to a common reference measure. Is it possible to show $(1)$ in general?
probability-theory measure-theory total-variation
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
$endgroup$
$begingroup$
Your guess is wrong. The equation fails when $mu$ and $nu$ are singular.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:15
$begingroup$
If you downvoted me please read my revised answer.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:25
$begingroup$
@KaviRamaMurthy Why did you delete your answer?
$endgroup$
– 0xbadf00d
Feb 25 at 17:48
add a comment |
$begingroup$
Let $E$ be a set and $mathcal Esubseteq 2^E$ with $emptysetinmathcal E$. If $mu:mathcal Etomathbb R$ with $mu(emptyset)=0$, then $$operatorname{Var}_mu(B):=supleft{sum_{i=1}^nleft|mu(B_i)right|:ninmathbb Ntext{ and }B_1,ldots,B_ninmathcal Btext{ are disjoint with }biguplus_{i=1}^nB_isubseteq Bright}$$ for $Bsubseteq E$. If $Einmathcal E$, let $$left|muright|:=operatorname{Var}_mu(E).$$
Assume $(E,mathcal E)$ is a measurable space and $mu,nu$ are probability measures on $(E,mathcal E)$. Is it possible to show that $$left|mu-nuright|=sup_{Binmathcal E}left|mu(B)-nu(B)right|tag1$$ or is there a counterexample?
I was able to show the claim assuming that $mu$ and $nu$ are both absolutely continuous with respect to a common reference measure. Is it possible to show $(1)$ in general?
probability-theory measure-theory total-variation
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
$endgroup$
Let $E$ be a set and $mathcal Esubseteq 2^E$ with $emptysetinmathcal E$. If $mu:mathcal Etomathbb R$ with $mu(emptyset)=0$, then $$operatorname{Var}_mu(B):=supleft{sum_{i=1}^nleft|mu(B_i)right|:ninmathbb Ntext{ and }B_1,ldots,B_ninmathcal Btext{ are disjoint with }biguplus_{i=1}^nB_isubseteq Bright}$$ for $Bsubseteq E$. If $Einmathcal E$, let $$left|muright|:=operatorname{Var}_mu(E).$$
Assume $(E,mathcal E)$ is a measurable space and $mu,nu$ are probability measures on $(E,mathcal E)$. Is it possible to show that $$left|mu-nuright|=sup_{Binmathcal E}left|mu(B)-nu(B)right|tag1$$ or is there a counterexample?
I was able to show the claim assuming that $mu$ and $nu$ are both absolutely continuous with respect to a common reference measure. Is it possible to show $(1)$ in general?
probability-theory measure-theory total-variation
probability-theory measure-theory total-variation
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
asked Dec 29 '18 at 11:57
0xbadf00d0xbadf00d
1,85941533
1,85941533
$begingroup$
Your guess is wrong. The equation fails when $mu$ and $nu$ are singular.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:15
$begingroup$
If you downvoted me please read my revised answer.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:25
$begingroup$
@KaviRamaMurthy Why did you delete your answer?
$endgroup$
– 0xbadf00d
Feb 25 at 17:48
add a comment |
$begingroup$
Your guess is wrong. The equation fails when $mu$ and $nu$ are singular.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:15
$begingroup$
If you downvoted me please read my revised answer.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:25
$begingroup$
@KaviRamaMurthy Why did you delete your answer?
$endgroup$
– 0xbadf00d
Feb 25 at 17:48
$begingroup$
Your guess is wrong. The equation fails when $mu$ and $nu$ are singular.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:15
$begingroup$
Your guess is wrong. The equation fails when $mu$ and $nu$ are singular.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:15
$begingroup$
If you downvoted me please read my revised answer.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:25
$begingroup$
If you downvoted me please read my revised answer.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:25
$begingroup$
@KaviRamaMurthy Why did you delete your answer?
$endgroup$
– 0xbadf00d
Feb 25 at 17:48
$begingroup$
@KaviRamaMurthy Why did you delete your answer?
$endgroup$
– 0xbadf00d
Feb 25 at 17:48
add a comment |
2 Answers
2
active
oldest
votes
$begingroup$
You are off by a factor of 2: $| mu-nu| =2sup_{Binmathcal E}|mu(B)-nu(B)|$. For this it's important that $mu $ and $nu$ are probabilities, so that $mu(E)-nu(E)=0$.
Later addition: Let $lambda=mu+nu$ and let $f$ and $g$ be the Radon-Nikodym densities of $mu$ and $nu$ with respect to $lambda$. It's not hard to check that $|mu-nu|=int_E|f-g|,dlambda$.
First, because $mu(E)=nu(E)=1$, if $Binmathcal E$ then,
$$
2|mu(B)-nu(B)|=|mu(B)-nu(B)|+|mu(B^c)-nu(B^c)|leoperatorname{Var}_{mu-nu}(E)=|mu-nu|.
$$
This shows that $|mu-nu|ge 2sup_{Binmathcal E}|mu(B)-nu(B)|$.
For the reverse inequality,
$$
eqalign{
|mu-nu|&=int_E|f-g|,dlambdacr
&=int_{{f>g}}(f-g),dlambda+int_{{f<g}}(g-f),dlambdacr
&=(mu-nu)({f>g})+(nu-mu)({f<g})cr
&=|(mu-nu)({f>g})|+|(mu-nu)({f<g})|cr
&le 2sup_{Binmathcal E}|mu(B)-nu(B)|.
}
$$
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
$endgroup$
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
add a comment |
$begingroup$
Your claim that the equality holds when the measures are absolutely continuous w.r.t. some measure is wrong. You can check that it fails even in simple cases like $mu (A)=m(Acap (0,1))$, $nu (A)= m(Acap (-1,0))$ where $m$ is Lebesgue measure. [Any two probability measures are absolutely continuous w.r.t some probability measure, e.g. $frac {mu +nu} 2$].
If $mu =delta_a$ and $nu =delta_b$ with $a neq b$ then $|mu-nu|=2$ and $|mu(B)-nu(B)| leq 1$ for all $B$.
[$|mu-nu| geq |(mu-nu)({a})|+|(mu-nu)({b})|=2$].
More generally if $P$ and $Q$ are any two probability measures with $Pperp Q$ the $|P-Q|=2$ and $|P(B)-Q(B)| leq 1$ for all $B$.
What is true in general is RHS $leq$ LHS $leq 2$RHS.
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
$endgroup$
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
add a comment |
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
});
}
});
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3055773%2ftotal-variation-distance-of-probaiblity-measures%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
2 Answers
2
active
oldest
votes
2 Answers
2
active
oldest
votes
active
oldest
votes
active
oldest
votes
$begingroup$
You are off by a factor of 2: $| mu-nu| =2sup_{Binmathcal E}|mu(B)-nu(B)|$. For this it's important that $mu $ and $nu$ are probabilities, so that $mu(E)-nu(E)=0$.
Later addition: Let $lambda=mu+nu$ and let $f$ and $g$ be the Radon-Nikodym densities of $mu$ and $nu$ with respect to $lambda$. It's not hard to check that $|mu-nu|=int_E|f-g|,dlambda$.
First, because $mu(E)=nu(E)=1$, if $Binmathcal E$ then,
$$
2|mu(B)-nu(B)|=|mu(B)-nu(B)|+|mu(B^c)-nu(B^c)|leoperatorname{Var}_{mu-nu}(E)=|mu-nu|.
$$
This shows that $|mu-nu|ge 2sup_{Binmathcal E}|mu(B)-nu(B)|$.
For the reverse inequality,
$$
eqalign{
|mu-nu|&=int_E|f-g|,dlambdacr
&=int_{{f>g}}(f-g),dlambda+int_{{f<g}}(g-f),dlambdacr
&=(mu-nu)({f>g})+(nu-mu)({f<g})cr
&=|(mu-nu)({f>g})|+|(mu-nu)({f<g})|cr
&le 2sup_{Binmathcal E}|mu(B)-nu(B)|.
}
$$
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
$endgroup$
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
add a comment |
$begingroup$
You are off by a factor of 2: $| mu-nu| =2sup_{Binmathcal E}|mu(B)-nu(B)|$. For this it's important that $mu $ and $nu$ are probabilities, so that $mu(E)-nu(E)=0$.
Later addition: Let $lambda=mu+nu$ and let $f$ and $g$ be the Radon-Nikodym densities of $mu$ and $nu$ with respect to $lambda$. It's not hard to check that $|mu-nu|=int_E|f-g|,dlambda$.
First, because $mu(E)=nu(E)=1$, if $Binmathcal E$ then,
$$
2|mu(B)-nu(B)|=|mu(B)-nu(B)|+|mu(B^c)-nu(B^c)|leoperatorname{Var}_{mu-nu}(E)=|mu-nu|.
$$
This shows that $|mu-nu|ge 2sup_{Binmathcal E}|mu(B)-nu(B)|$.
For the reverse inequality,
$$
eqalign{
|mu-nu|&=int_E|f-g|,dlambdacr
&=int_{{f>g}}(f-g),dlambda+int_{{f<g}}(g-f),dlambdacr
&=(mu-nu)({f>g})+(nu-mu)({f<g})cr
&=|(mu-nu)({f>g})|+|(mu-nu)({f<g})|cr
&le 2sup_{Binmathcal E}|mu(B)-nu(B)|.
}
$$
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
$endgroup$
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
add a comment |
$begingroup$
You are off by a factor of 2: $| mu-nu| =2sup_{Binmathcal E}|mu(B)-nu(B)|$. For this it's important that $mu $ and $nu$ are probabilities, so that $mu(E)-nu(E)=0$.
Later addition: Let $lambda=mu+nu$ and let $f$ and $g$ be the Radon-Nikodym densities of $mu$ and $nu$ with respect to $lambda$. It's not hard to check that $|mu-nu|=int_E|f-g|,dlambda$.
First, because $mu(E)=nu(E)=1$, if $Binmathcal E$ then,
$$
2|mu(B)-nu(B)|=|mu(B)-nu(B)|+|mu(B^c)-nu(B^c)|leoperatorname{Var}_{mu-nu}(E)=|mu-nu|.
$$
This shows that $|mu-nu|ge 2sup_{Binmathcal E}|mu(B)-nu(B)|$.
For the reverse inequality,
$$
eqalign{
|mu-nu|&=int_E|f-g|,dlambdacr
&=int_{{f>g}}(f-g),dlambda+int_{{f<g}}(g-f),dlambdacr
&=(mu-nu)({f>g})+(nu-mu)({f<g})cr
&=|(mu-nu)({f>g})|+|(mu-nu)({f<g})|cr
&le 2sup_{Binmathcal E}|mu(B)-nu(B)|.
}
$$
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
$endgroup$
You are off by a factor of 2: $| mu-nu| =2sup_{Binmathcal E}|mu(B)-nu(B)|$. For this it's important that $mu $ and $nu$ are probabilities, so that $mu(E)-nu(E)=0$.
Later addition: Let $lambda=mu+nu$ and let $f$ and $g$ be the Radon-Nikodym densities of $mu$ and $nu$ with respect to $lambda$. It's not hard to check that $|mu-nu|=int_E|f-g|,dlambda$.
First, because $mu(E)=nu(E)=1$, if $Binmathcal E$ then,
$$
2|mu(B)-nu(B)|=|mu(B)-nu(B)|+|mu(B^c)-nu(B^c)|leoperatorname{Var}_{mu-nu}(E)=|mu-nu|.
$$
This shows that $|mu-nu|ge 2sup_{Binmathcal E}|mu(B)-nu(B)|$.
For the reverse inequality,
$$
eqalign{
|mu-nu|&=int_E|f-g|,dlambdacr
&=int_{{f>g}}(f-g),dlambda+int_{{f<g}}(g-f),dlambdacr
&=(mu-nu)({f>g})+(nu-mu)({f<g})cr
&=|(mu-nu)({f>g})|+|(mu-nu)({f<g})|cr
&le 2sup_{Binmathcal E}|mu(B)-nu(B)|.
}
$$
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
edited Feb 25 at 20:06
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
answered Dec 29 '18 at 18:30
John DawkinsJohn Dawkins
13.3k11017
13.3k11017
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
add a comment |
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
Maybe I forgot the factor $2$, but how is $mu(E)-mu(E)=1-1=0$ answering my question? Could you provide a proof for the identity? (I didn't downvote your answer.)
$endgroup$
– 0xbadf00d
Dec 30 '18 at 0:16
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
It answers your question in the sense that it points out that you have the wrong formula except in the trivial case when both sides of the = sign are 0. So the answer to your question, put plainly, is NO.
$endgroup$
– John Dawkins
Jan 4 at 23:04
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
Yes, okay. I didn't ask explicitly how we can prove the correct formula if mine would turn out to be wrong. However, can you tell me how we can prove it?
$endgroup$
– 0xbadf00d
Feb 24 at 18:19
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
$begingroup$
I've added a proof to my original answer.
$endgroup$
– John Dawkins
Feb 25 at 20:06
add a comment |
$begingroup$
Your claim that the equality holds when the measures are absolutely continuous w.r.t. some measure is wrong. You can check that it fails even in simple cases like $mu (A)=m(Acap (0,1))$, $nu (A)= m(Acap (-1,0))$ where $m$ is Lebesgue measure. [Any two probability measures are absolutely continuous w.r.t some probability measure, e.g. $frac {mu +nu} 2$].
If $mu =delta_a$ and $nu =delta_b$ with $a neq b$ then $|mu-nu|=2$ and $|mu(B)-nu(B)| leq 1$ for all $B$.
[$|mu-nu| geq |(mu-nu)({a})|+|(mu-nu)({b})|=2$].
More generally if $P$ and $Q$ are any two probability measures with $Pperp Q$ the $|P-Q|=2$ and $|P(B)-Q(B)| leq 1$ for all $B$.
What is true in general is RHS $leq$ LHS $leq 2$RHS.
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
$endgroup$
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
add a comment |
$begingroup$
Your claim that the equality holds when the measures are absolutely continuous w.r.t. some measure is wrong. You can check that it fails even in simple cases like $mu (A)=m(Acap (0,1))$, $nu (A)= m(Acap (-1,0))$ where $m$ is Lebesgue measure. [Any two probability measures are absolutely continuous w.r.t some probability measure, e.g. $frac {mu +nu} 2$].
If $mu =delta_a$ and $nu =delta_b$ with $a neq b$ then $|mu-nu|=2$ and $|mu(B)-nu(B)| leq 1$ for all $B$.
[$|mu-nu| geq |(mu-nu)({a})|+|(mu-nu)({b})|=2$].
More generally if $P$ and $Q$ are any two probability measures with $Pperp Q$ the $|P-Q|=2$ and $|P(B)-Q(B)| leq 1$ for all $B$.
What is true in general is RHS $leq$ LHS $leq 2$RHS.
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
$endgroup$
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
add a comment |
$begingroup$
Your claim that the equality holds when the measures are absolutely continuous w.r.t. some measure is wrong. You can check that it fails even in simple cases like $mu (A)=m(Acap (0,1))$, $nu (A)= m(Acap (-1,0))$ where $m$ is Lebesgue measure. [Any two probability measures are absolutely continuous w.r.t some probability measure, e.g. $frac {mu +nu} 2$].
If $mu =delta_a$ and $nu =delta_b$ with $a neq b$ then $|mu-nu|=2$ and $|mu(B)-nu(B)| leq 1$ for all $B$.
[$|mu-nu| geq |(mu-nu)({a})|+|(mu-nu)({b})|=2$].
More generally if $P$ and $Q$ are any two probability measures with $Pperp Q$ the $|P-Q|=2$ and $|P(B)-Q(B)| leq 1$ for all $B$.
What is true in general is RHS $leq$ LHS $leq 2$RHS.
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
$endgroup$
Your claim that the equality holds when the measures are absolutely continuous w.r.t. some measure is wrong. You can check that it fails even in simple cases like $mu (A)=m(Acap (0,1))$, $nu (A)= m(Acap (-1,0))$ where $m$ is Lebesgue measure. [Any two probability measures are absolutely continuous w.r.t some probability measure, e.g. $frac {mu +nu} 2$].
If $mu =delta_a$ and $nu =delta_b$ with $a neq b$ then $|mu-nu|=2$ and $|mu(B)-nu(B)| leq 1$ for all $B$.
[$|mu-nu| geq |(mu-nu)({a})|+|(mu-nu)({b})|=2$].
More generally if $P$ and $Q$ are any two probability measures with $Pperp Q$ the $|P-Q|=2$ and $|P(B)-Q(B)| leq 1$ for all $B$.
What is true in general is RHS $leq$ LHS $leq 2$RHS.
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
edited Dec 30 '18 at 4:40
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
answered Dec 29 '18 at 12:09
Kavi Rama MurthyKavi Rama Murthy
68.6k53169
68.6k53169
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
add a comment |
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can someone tell me what is wrong with this answer?
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:13
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
$begingroup$
Can we even show that "$text{LHS}=2text{RHS}$"?
$endgroup$
– 0xbadf00d
Feb 24 at 19:27
add a comment |
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.
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3055773%2ftotal-variation-distance-of-probaiblity-measures%23new-answer', 'question_page');
}
);
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Post as a guest
Required, but never shown
Sign up or log in
StackExchange.ready(function () {
StackExchange.helpers.onClickDraftSave('#login-link');
});
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
Sign up using Google
Sign up using Facebook
Sign up using Email and Password
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
$begingroup$
Your guess is wrong. The equation fails when $mu$ and $nu$ are singular.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:15
$begingroup$
If you downvoted me please read my revised answer.
$endgroup$
– Kavi Rama Murthy
Dec 29 '18 at 23:25
$begingroup$
@KaviRamaMurthy Why did you delete your answer?
$endgroup$
– 0xbadf00d
Feb 25 at 17:48