Matlab : best approach for estimation
.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty{ height:90px;width:728px;box-sizing:border-box;
}
I would like to estimate parameters :
I have a PSF (Point Spread Function) "y=f(x)" depending of 6 parameters
I have to estimate all the 6 parameters (a,b,r0,c0,alpha,beta) given only "y" output data.
What is the best approach from a statistical point of view ? Can I find the estimations with Matlab fminsearch function ?
Thanks
matlab
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
add a comment |
I would like to estimate parameters :
I have a PSF (Point Spread Function) "y=f(x)" depending of 6 parameters
I have to estimate all the 6 parameters (a,b,r0,c0,alpha,beta) given only "y" output data.
What is the best approach from a statistical point of view ? Can I find the estimations with Matlab fminsearch function ?
Thanks
matlab
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
For thewhilevsforloop: I like to use a hybrid:kmax = 1e6; while Jmin>epsilon && k<kmax;and then incrementkeach iteration. This way it will stop when the result is good enough, and if it doesn't reach 'good enough' within the specified number of iterations, it'll stop anyway.
– Adriaan
Nov 16 '18 at 13:13
-@Adriaan thanks, I just put your suggestion
– youpilat13
Nov 16 '18 at 13:29
See the edit to my answer
– rinkert
Nov 16 '18 at 14:55
-@rinkert the results withfminsearch,fminconandgadon't give significant results compared to the parameters fixed before the estimation.There should be an error but I can't find it. Could you take a look please at my source located in the archive on : 31.207.34.24/Archive_Estimation_SO.tar. You will find the cost function inCrit_J.mand the main source istp_estimation_dev.m. Right now, I can't explain this bad estimation of the six parameters that I want to get. Regards
– youpilat13
Nov 16 '18 at 17:14
add a comment |
I would like to estimate parameters :
I have a PSF (Point Spread Function) "y=f(x)" depending of 6 parameters
I have to estimate all the 6 parameters (a,b,r0,c0,alpha,beta) given only "y" output data.
What is the best approach from a statistical point of view ? Can I find the estimations with Matlab fminsearch function ?
Thanks
matlab
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
I would like to estimate parameters :
I have a PSF (Point Spread Function) "y=f(x)" depending of 6 parameters
I have to estimate all the 6 parameters (a,b,r0,c0,alpha,beta) given only "y" output data.
What is the best approach from a statistical point of view ? Can I find the estimations with Matlab fminsearch function ?
Thanks
matlab
matlab
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
edited Nov 17 '18 at 15:28
youpilat13
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
asked Nov 16 '18 at 13:08
youpilat13youpilat13
5221442
5221442
For thewhilevsforloop: I like to use a hybrid:kmax = 1e6; while Jmin>epsilon && k<kmax;and then incrementkeach iteration. This way it will stop when the result is good enough, and if it doesn't reach 'good enough' within the specified number of iterations, it'll stop anyway.
– Adriaan
Nov 16 '18 at 13:13
-@Adriaan thanks, I just put your suggestion
– youpilat13
Nov 16 '18 at 13:29
See the edit to my answer
– rinkert
Nov 16 '18 at 14:55
-@rinkert the results withfminsearch,fminconandgadon't give significant results compared to the parameters fixed before the estimation.There should be an error but I can't find it. Could you take a look please at my source located in the archive on : 31.207.34.24/Archive_Estimation_SO.tar. You will find the cost function inCrit_J.mand the main source istp_estimation_dev.m. Right now, I can't explain this bad estimation of the six parameters that I want to get. Regards
– youpilat13
Nov 16 '18 at 17:14
add a comment |
For thewhilevsforloop: I like to use a hybrid:kmax = 1e6; while Jmin>epsilon && k<kmax;and then incrementkeach iteration. This way it will stop when the result is good enough, and if it doesn't reach 'good enough' within the specified number of iterations, it'll stop anyway.
– Adriaan
Nov 16 '18 at 13:13
-@Adriaan thanks, I just put your suggestion
– youpilat13
Nov 16 '18 at 13:29
See the edit to my answer
– rinkert
Nov 16 '18 at 14:55
-@rinkert the results withfminsearch,fminconandgadon't give significant results compared to the parameters fixed before the estimation.There should be an error but I can't find it. Could you take a look please at my source located in the archive on : 31.207.34.24/Archive_Estimation_SO.tar. You will find the cost function inCrit_J.mand the main source istp_estimation_dev.m. Right now, I can't explain this bad estimation of the six parameters that I want to get. Regards
– youpilat13
Nov 16 '18 at 17:14
For the
while vs for loop: I like to use a hybrid: kmax = 1e6; while Jmin>epsilon && k<kmax; and then increment k each iteration. This way it will stop when the result is good enough, and if it doesn't reach 'good enough' within the specified number of iterations, it'll stop anyway.– Adriaan
Nov 16 '18 at 13:13
For the
while vs for loop: I like to use a hybrid: kmax = 1e6; while Jmin>epsilon && k<kmax; and then increment k each iteration. This way it will stop when the result is good enough, and if it doesn't reach 'good enough' within the specified number of iterations, it'll stop anyway.– Adriaan
Nov 16 '18 at 13:13
-@Adriaan thanks, I just put your suggestion
– youpilat13
Nov 16 '18 at 13:29
-@Adriaan thanks, I just put your suggestion
– youpilat13
Nov 16 '18 at 13:29
See the edit to my answer
– rinkert
Nov 16 '18 at 14:55
See the edit to my answer
– rinkert
Nov 16 '18 at 14:55
-@rinkert the results with
fminsearch, fmincon and ga don't give significant results compared to the parameters fixed before the estimation.There should be an error but I can't find it. Could you take a look please at my source located in the archive on : 31.207.34.24/Archive_Estimation_SO.tar. You will find the cost function in Crit_J.mand the main source is tp_estimation_dev.m . Right now, I can't explain this bad estimation of the six parameters that I want to get. Regards– youpilat13
Nov 16 '18 at 17:14
-@rinkert the results with
fminsearch, fmincon and ga don't give significant results compared to the parameters fixed before the estimation.There should be an error but I can't find it. Could you take a look please at my source located in the archive on : 31.207.34.24/Archive_Estimation_SO.tar. You will find the cost function in Crit_J.mand the main source is tp_estimation_dev.m . Right now, I can't explain this bad estimation of the six parameters that I want to get. Regards– youpilat13
Nov 16 '18 at 17:14
add a comment |
1 Answer
1
active
oldest
votes
Right now you are just randomly changing your parameters. If you change to an optimization procedure like implemented in fminsearch, this algorithm will find the minimum for you.
% initial parameters
a = 10;
b = 3;
r0 = 0;
c0 = 0;
alpha = 1;
beta = 1;
% Array of parameters
p0 = [a,b,r0,c0,alpha,beta]';
% Minimization of cost function for p
[pars, Jmin] = fminsearch(@(x)Crit_J(x,D), p0);
You have to change your Crit_J function to only output the 'cost' to minimize for this to work. You can then later obtain your model by running the original function.
Update:
Perhaps you are finding a local minimum, you can try fminunc instead of fminsearch, or switch to a generic algorithm, but that might be overkill for this problem.
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
add a comment |
Your Answer
StackExchange.ifUsing("editor", function () {
StackExchange.using("externalEditor", function () {
StackExchange.using("snippets", function () {
StackExchange.snippets.init();
});
});
}, "code-snippets");
StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "1"
};
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
},
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%2fstackoverflow.com%2fquestions%2f53338554%2fmatlab-best-approach-for-estimation%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
Right now you are just randomly changing your parameters. If you change to an optimization procedure like implemented in fminsearch, this algorithm will find the minimum for you.
% initial parameters
a = 10;
b = 3;
r0 = 0;
c0 = 0;
alpha = 1;
beta = 1;
% Array of parameters
p0 = [a,b,r0,c0,alpha,beta]';
% Minimization of cost function for p
[pars, Jmin] = fminsearch(@(x)Crit_J(x,D), p0);
You have to change your Crit_J function to only output the 'cost' to minimize for this to work. You can then later obtain your model by running the original function.
Update:
Perhaps you are finding a local minimum, you can try fminunc instead of fminsearch, or switch to a generic algorithm, but that might be overkill for this problem.
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
add a comment |
Right now you are just randomly changing your parameters. If you change to an optimization procedure like implemented in fminsearch, this algorithm will find the minimum for you.
% initial parameters
a = 10;
b = 3;
r0 = 0;
c0 = 0;
alpha = 1;
beta = 1;
% Array of parameters
p0 = [a,b,r0,c0,alpha,beta]';
% Minimization of cost function for p
[pars, Jmin] = fminsearch(@(x)Crit_J(x,D), p0);
You have to change your Crit_J function to only output the 'cost' to minimize for this to work. You can then later obtain your model by running the original function.
Update:
Perhaps you are finding a local minimum, you can try fminunc instead of fminsearch, or switch to a generic algorithm, but that might be overkill for this problem.
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
add a comment |
Right now you are just randomly changing your parameters. If you change to an optimization procedure like implemented in fminsearch, this algorithm will find the minimum for you.
% initial parameters
a = 10;
b = 3;
r0 = 0;
c0 = 0;
alpha = 1;
beta = 1;
% Array of parameters
p0 = [a,b,r0,c0,alpha,beta]';
% Minimization of cost function for p
[pars, Jmin] = fminsearch(@(x)Crit_J(x,D), p0);
You have to change your Crit_J function to only output the 'cost' to minimize for this to work. You can then later obtain your model by running the original function.
Update:
Perhaps you are finding a local minimum, you can try fminunc instead of fminsearch, or switch to a generic algorithm, but that might be overkill for this problem.
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
Right now you are just randomly changing your parameters. If you change to an optimization procedure like implemented in fminsearch, this algorithm will find the minimum for you.
% initial parameters
a = 10;
b = 3;
r0 = 0;
c0 = 0;
alpha = 1;
beta = 1;
% Array of parameters
p0 = [a,b,r0,c0,alpha,beta]';
% Minimization of cost function for p
[pars, Jmin] = fminsearch(@(x)Crit_J(x,D), p0);
You have to change your Crit_J function to only output the 'cost' to minimize for this to work. You can then later obtain your model by running the original function.
Update:
Perhaps you are finding a local minimum, you can try fminunc instead of fminsearch, or switch to a generic algorithm, but that might be overkill for this problem.
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
edited Nov 16 '18 at 14:54
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
answered Nov 16 '18 at 13:29
rinkertrinkert
1,5731618
1,5731618
add a comment |
add a comment |
Thanks for contributing an answer to Stack Overflow!
- 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.
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%2fstackoverflow.com%2fquestions%2f53338554%2fmatlab-best-approach-for-estimation%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
For the
whilevsforloop: I like to use a hybrid:kmax = 1e6; while Jmin>epsilon && k<kmax;and then incrementkeach iteration. This way it will stop when the result is good enough, and if it doesn't reach 'good enough' within the specified number of iterations, it'll stop anyway.– Adriaan
Nov 16 '18 at 13:13
-@Adriaan thanks, I just put your suggestion
– youpilat13
Nov 16 '18 at 13:29
See the edit to my answer
– rinkert
Nov 16 '18 at 14:55
-@rinkert the results with
fminsearch,fminconandgadon't give significant results compared to the parameters fixed before the estimation.There should be an error but I can't find it. Could you take a look please at my source located in the archive on : 31.207.34.24/Archive_Estimation_SO.tar. You will find the cost function inCrit_J.mand the main source istp_estimation_dev.m. Right now, I can't explain this bad estimation of the six parameters that I want to get. Regards– youpilat13
Nov 16 '18 at 17:14