Harmonic functions and Markov processes
Clash Royale CLAN TAG#URR8PPP
up vote
0
down vote
favorite
I was reading some lecture notes, in which it is stated that, given a discrete time Markov process $X_t$, for its Markov semigroup $P_t$ it holds that:
$$
P_tf(x) = E_x[f(X_t)]=E[f(X_t)|X_0=x]
$$
For each $f:S to mathbfR$ measurable and bounded. After this statement it is said that a bounded measurable function $f:Sto mathbfR$ (where S is a Polish space endowed with its Borel ÃÂ-field) is said to be harmonic on DâÂÂS iff
$$ Lf(x)=0 forall x in D$$
where $ L=PâÂÂ1$ is the generator of the markov process. It is asked to prove that such a function $f $ is harmonic on $S$ if and only if the process $ f(X_t) $ is a martingale. I tried to prove it, and it is easy to see, using the fact that
$$M_t:=f(X_t)âÂÂf(X_0)âÂÂsum_s=0^t-1Lf(X_s)$$
is a martingale, that $f(X_t)$ is a martingale if and only if $Lf(X_s)=0 forall s ge 0$. Now, if f is harmonic it is clear that $Lf(X_s)=0$, for the converse it seems that one should derive the fact that $ f $ is harmonic from$ Lf(X_0)=0$, indeed one has
$$0=Lf(X_0)=Pf(X_0)âÂÂf(X_0)$$
now, I was able to prove the following:
If $ g:SâÂÂmathbfR$ is bounded and measurable, and such that
$$g(X_0)=E[f(X_t)|F_0]$$
then$ g(x)=E[f(Xt)|X_0=x]$, $P_X_0$-a.e.
Since we know that $ E[f(X_1)|X_0=x]=Pf(x)$, one gets that $P_X_0$-a.e.
$$f(x)=Pf(x)$$
and thus$ Lf(x)=0$, $P_X_0$-a.e. But I'm not able to prove that equality holds for every x. Do you have any idea?
probability markov-process
This question has an open bounty worth +50
reputation from jJjjJ ending ending at 2018-09-04 12:31:53Z">in 4 days.
The question is widely applicable to a large audience. A detailed canonical answer is required to address all the concerns.
add a comment |Â
up vote
0
down vote
favorite
I was reading some lecture notes, in which it is stated that, given a discrete time Markov process $X_t$, for its Markov semigroup $P_t$ it holds that:
$$
P_tf(x) = E_x[f(X_t)]=E[f(X_t)|X_0=x]
$$
For each $f:S to mathbfR$ measurable and bounded. After this statement it is said that a bounded measurable function $f:Sto mathbfR$ (where S is a Polish space endowed with its Borel ÃÂ-field) is said to be harmonic on DâÂÂS iff
$$ Lf(x)=0 forall x in D$$
where $ L=PâÂÂ1$ is the generator of the markov process. It is asked to prove that such a function $f $ is harmonic on $S$ if and only if the process $ f(X_t) $ is a martingale. I tried to prove it, and it is easy to see, using the fact that
$$M_t:=f(X_t)âÂÂf(X_0)âÂÂsum_s=0^t-1Lf(X_s)$$
is a martingale, that $f(X_t)$ is a martingale if and only if $Lf(X_s)=0 forall s ge 0$. Now, if f is harmonic it is clear that $Lf(X_s)=0$, for the converse it seems that one should derive the fact that $ f $ is harmonic from$ Lf(X_0)=0$, indeed one has
$$0=Lf(X_0)=Pf(X_0)âÂÂf(X_0)$$
now, I was able to prove the following:
If $ g:SâÂÂmathbfR$ is bounded and measurable, and such that
$$g(X_0)=E[f(X_t)|F_0]$$
then$ g(x)=E[f(Xt)|X_0=x]$, $P_X_0$-a.e.
Since we know that $ E[f(X_1)|X_0=x]=Pf(x)$, one gets that $P_X_0$-a.e.
$$f(x)=Pf(x)$$
and thus$ Lf(x)=0$, $P_X_0$-a.e. But I'm not able to prove that equality holds for every x. Do you have any idea?
probability markov-process
This question has an open bounty worth +50
reputation from jJjjJ ending ending at 2018-09-04 12:31:53Z">in 4 days.
The question is widely applicable to a large audience. A detailed canonical answer is required to address all the concerns.
add a comment |Â
up vote
0
down vote
favorite
up vote
0
down vote
favorite
I was reading some lecture notes, in which it is stated that, given a discrete time Markov process $X_t$, for its Markov semigroup $P_t$ it holds that:
$$
P_tf(x) = E_x[f(X_t)]=E[f(X_t)|X_0=x]
$$
For each $f:S to mathbfR$ measurable and bounded. After this statement it is said that a bounded measurable function $f:Sto mathbfR$ (where S is a Polish space endowed with its Borel ÃÂ-field) is said to be harmonic on DâÂÂS iff
$$ Lf(x)=0 forall x in D$$
where $ L=PâÂÂ1$ is the generator of the markov process. It is asked to prove that such a function $f $ is harmonic on $S$ if and only if the process $ f(X_t) $ is a martingale. I tried to prove it, and it is easy to see, using the fact that
$$M_t:=f(X_t)âÂÂf(X_0)âÂÂsum_s=0^t-1Lf(X_s)$$
is a martingale, that $f(X_t)$ is a martingale if and only if $Lf(X_s)=0 forall s ge 0$. Now, if f is harmonic it is clear that $Lf(X_s)=0$, for the converse it seems that one should derive the fact that $ f $ is harmonic from$ Lf(X_0)=0$, indeed one has
$$0=Lf(X_0)=Pf(X_0)âÂÂf(X_0)$$
now, I was able to prove the following:
If $ g:SâÂÂmathbfR$ is bounded and measurable, and such that
$$g(X_0)=E[f(X_t)|F_0]$$
then$ g(x)=E[f(Xt)|X_0=x]$, $P_X_0$-a.e.
Since we know that $ E[f(X_1)|X_0=x]=Pf(x)$, one gets that $P_X_0$-a.e.
$$f(x)=Pf(x)$$
and thus$ Lf(x)=0$, $P_X_0$-a.e. But I'm not able to prove that equality holds for every x. Do you have any idea?
probability markov-process
I was reading some lecture notes, in which it is stated that, given a discrete time Markov process $X_t$, for its Markov semigroup $P_t$ it holds that:
$$
P_tf(x) = E_x[f(X_t)]=E[f(X_t)|X_0=x]
$$
For each $f:S to mathbfR$ measurable and bounded. After this statement it is said that a bounded measurable function $f:Sto mathbfR$ (where S is a Polish space endowed with its Borel ÃÂ-field) is said to be harmonic on DâÂÂS iff
$$ Lf(x)=0 forall x in D$$
where $ L=PâÂÂ1$ is the generator of the markov process. It is asked to prove that such a function $f $ is harmonic on $S$ if and only if the process $ f(X_t) $ is a martingale. I tried to prove it, and it is easy to see, using the fact that
$$M_t:=f(X_t)âÂÂf(X_0)âÂÂsum_s=0^t-1Lf(X_s)$$
is a martingale, that $f(X_t)$ is a martingale if and only if $Lf(X_s)=0 forall s ge 0$. Now, if f is harmonic it is clear that $Lf(X_s)=0$, for the converse it seems that one should derive the fact that $ f $ is harmonic from$ Lf(X_0)=0$, indeed one has
$$0=Lf(X_0)=Pf(X_0)âÂÂf(X_0)$$
now, I was able to prove the following:
If $ g:SâÂÂmathbfR$ is bounded and measurable, and such that
$$g(X_0)=E[f(X_t)|F_0]$$
then$ g(x)=E[f(Xt)|X_0=x]$, $P_X_0$-a.e.
Since we know that $ E[f(X_1)|X_0=x]=Pf(x)$, one gets that $P_X_0$-a.e.
$$f(x)=Pf(x)$$
and thus$ Lf(x)=0$, $P_X_0$-a.e. But I'm not able to prove that equality holds for every x. Do you have any idea?
probability markov-process
edited Aug 19 at 12:31
asked Aug 18 at 6:44
jJjjJ
324311
324311
This question has an open bounty worth +50
reputation from jJjjJ ending ending at 2018-09-04 12:31:53Z">in 4 days.
The question is widely applicable to a large audience. A detailed canonical answer is required to address all the concerns.
This question has an open bounty worth +50
reputation from jJjjJ ending ending at 2018-09-04 12:31:53Z">in 4 days.
The question is widely applicable to a large audience. A detailed canonical answer is required to address all the concerns.
add a comment |Â
add a comment |Â
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
active
oldest
votes
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
StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2886481%2fharmonic-functions-and-markov-processes%23new-answer', 'question_page');
);
Post as a guest
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
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
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