An unbounded function with integrability for all p>0

The name of the pictureThe name of the pictureThe name of the pictureClash Royale CLAN TAG#URR8PPP











up vote
3
down vote

favorite












How to construct a function $fgeq0$ such that i) $fin L^p([0,1])forall p>0$ and ii) $esssup_I f=+infty$ for any interval $Isubset[0,1]$ ?



Thank you!







share|cite|improve this question
























    up vote
    3
    down vote

    favorite












    How to construct a function $fgeq0$ such that i) $fin L^p([0,1])forall p>0$ and ii) $esssup_I f=+infty$ for any interval $Isubset[0,1]$ ?



    Thank you!







    share|cite|improve this question






















      up vote
      3
      down vote

      favorite









      up vote
      3
      down vote

      favorite











      How to construct a function $fgeq0$ such that i) $fin L^p([0,1])forall p>0$ and ii) $esssup_I f=+infty$ for any interval $Isubset[0,1]$ ?



      Thank you!







      share|cite|improve this question












      How to construct a function $fgeq0$ such that i) $fin L^p([0,1])forall p>0$ and ii) $esssup_I f=+infty$ for any interval $Isubset[0,1]$ ?



      Thank you!









      share|cite|improve this question











      share|cite|improve this question




      share|cite|improve this question










      asked Aug 29 at 10:31









      Leonardo

      1035




      1035




















          2 Answers
          2






          active

          oldest

          votes

















          up vote
          2
          down vote













          How about
          $$
          f(x)=|log x|?
          $$



          The above is a wrong answer, since it does not fulfill 3. So let's try again.



          If $0<a<1$, then
          $$
          int_0^1|log |x-a||,dx=1 - (1 - a) log(1 - a)- a log a<2.
          $$
          Order the rationals in $[0,1]$ in a sequence $a_n$, let $f_n(x)=bigl|,log |x-a_n|,bigr|$ and
          $$
          f(x)=sum_n=1^infty2^-nf_n(x) .
          $$
          Clearly $f$ is positive. Moreover
          $$
          int_0^1f(x),dx=sum_n=1^infty2^-nint_0^1f_n(x),dxle2 sum_n=1^infty2^-n<infty,
          $$
          so that $f$ is finite almost everywhere and in $L^1$. We have to check that $fin L^p$ for $p>1$. Since
          $$
          |f|_plesum_n=1^infty2^-n|f_n|_p,
          $$
          it is enough to show that $|f_n|_ple C_p$ for some constant $C_p$ depending only on $p$. The function $|x-a|^1/(2p)bigl|,log |x-a|bigr|$ is continuous, and in particular bounded, as a function of $(x,a)$ on $[0,1]times[0,1]$. It follows that there exists a constant $C_p$ such that
          $$
          bigl|,log |x-a|^pbigr|lefracC_psqrt,quad0le xle1,quad0le ale1,
          $$
          and the claim follows.






          share|cite|improve this answer






















          • Oops. Did not read the "any interval" part.
            – Julián Aguirre
            Aug 29 at 11:20










          • I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
            – Lorenzo Quarisa
            Aug 29 at 12:09











          • Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
            – Lorenzo Quarisa
            Aug 29 at 12:26










          • @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
            – Julián Aguirre
            Aug 29 at 13:45











          • @JuliánAguirre Oh, that is true also. Thanks.
            – Lorenzo Quarisa
            Aug 29 at 13:54

















          up vote
          1
          down vote













          Let
          $$mathbbQcap [0,1]=leftx_nright_nin mathbbN $$
          and
          $$f_n(x)=-log|x-x_n| $$
          so that $f_ngeq0$, $f_nin L^p(0,1)$ for all $pin (0,infty)$, and
          $$|f_n|_p^p=int_0^1|f_n|^p=int_0^1|log|x-x_n||^pdx leqint_-1^1|log |x||^pdx=:C<infty$$
          Let
          $$f=sum_nin mathbbNfrac12^nf_n $$
          then by Minkowski's inequality and Minkowski's inequality for $pin (0,1)$, we have
          $$|f|_pleq maxleft2^frac1p-1;1rightsum_nin mathbbNfrac12^n|f_n|_pleq max left2^frac1p-1;1rightC^1/psum_nin mathbbNfrac12^n<infty $$
          Hence $fin L^p$ for all $pin (0,infty)$, and for any interval $Isubset [0,1]$, there is $nin mathbbN$ such that $x_nin I$, and so $$operatornameess,supfgeq operatornameess,supf_n=+infty$$






          share|cite|improve this answer






















            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',
            convertImagesToLinks: true,
            noModals: false,
            showLowRepImageUploadWarning: true,
            reputationToPostImages: 10,
            bindNavPrevention: true,
            postfix: "",
            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%2f2898216%2fan-unbounded-function-with-integrability-for-all-p0%23new-answer', 'question_page');

            );

            Post as a guest






























            2 Answers
            2






            active

            oldest

            votes








            2 Answers
            2






            active

            oldest

            votes









            active

            oldest

            votes






            active

            oldest

            votes








            up vote
            2
            down vote













            How about
            $$
            f(x)=|log x|?
            $$



            The above is a wrong answer, since it does not fulfill 3. So let's try again.



            If $0<a<1$, then
            $$
            int_0^1|log |x-a||,dx=1 - (1 - a) log(1 - a)- a log a<2.
            $$
            Order the rationals in $[0,1]$ in a sequence $a_n$, let $f_n(x)=bigl|,log |x-a_n|,bigr|$ and
            $$
            f(x)=sum_n=1^infty2^-nf_n(x) .
            $$
            Clearly $f$ is positive. Moreover
            $$
            int_0^1f(x),dx=sum_n=1^infty2^-nint_0^1f_n(x),dxle2 sum_n=1^infty2^-n<infty,
            $$
            so that $f$ is finite almost everywhere and in $L^1$. We have to check that $fin L^p$ for $p>1$. Since
            $$
            |f|_plesum_n=1^infty2^-n|f_n|_p,
            $$
            it is enough to show that $|f_n|_ple C_p$ for some constant $C_p$ depending only on $p$. The function $|x-a|^1/(2p)bigl|,log |x-a|bigr|$ is continuous, and in particular bounded, as a function of $(x,a)$ on $[0,1]times[0,1]$. It follows that there exists a constant $C_p$ such that
            $$
            bigl|,log |x-a|^pbigr|lefracC_psqrt,quad0le xle1,quad0le ale1,
            $$
            and the claim follows.






            share|cite|improve this answer






















            • Oops. Did not read the "any interval" part.
              – Julián Aguirre
              Aug 29 at 11:20










            • I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
              – Lorenzo Quarisa
              Aug 29 at 12:09











            • Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
              – Lorenzo Quarisa
              Aug 29 at 12:26










            • @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
              – Julián Aguirre
              Aug 29 at 13:45











            • @JuliánAguirre Oh, that is true also. Thanks.
              – Lorenzo Quarisa
              Aug 29 at 13:54














            up vote
            2
            down vote













            How about
            $$
            f(x)=|log x|?
            $$



            The above is a wrong answer, since it does not fulfill 3. So let's try again.



            If $0<a<1$, then
            $$
            int_0^1|log |x-a||,dx=1 - (1 - a) log(1 - a)- a log a<2.
            $$
            Order the rationals in $[0,1]$ in a sequence $a_n$, let $f_n(x)=bigl|,log |x-a_n|,bigr|$ and
            $$
            f(x)=sum_n=1^infty2^-nf_n(x) .
            $$
            Clearly $f$ is positive. Moreover
            $$
            int_0^1f(x),dx=sum_n=1^infty2^-nint_0^1f_n(x),dxle2 sum_n=1^infty2^-n<infty,
            $$
            so that $f$ is finite almost everywhere and in $L^1$. We have to check that $fin L^p$ for $p>1$. Since
            $$
            |f|_plesum_n=1^infty2^-n|f_n|_p,
            $$
            it is enough to show that $|f_n|_ple C_p$ for some constant $C_p$ depending only on $p$. The function $|x-a|^1/(2p)bigl|,log |x-a|bigr|$ is continuous, and in particular bounded, as a function of $(x,a)$ on $[0,1]times[0,1]$. It follows that there exists a constant $C_p$ such that
            $$
            bigl|,log |x-a|^pbigr|lefracC_psqrt,quad0le xle1,quad0le ale1,
            $$
            and the claim follows.






            share|cite|improve this answer






















            • Oops. Did not read the "any interval" part.
              – Julián Aguirre
              Aug 29 at 11:20










            • I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
              – Lorenzo Quarisa
              Aug 29 at 12:09











            • Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
              – Lorenzo Quarisa
              Aug 29 at 12:26










            • @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
              – Julián Aguirre
              Aug 29 at 13:45











            • @JuliánAguirre Oh, that is true also. Thanks.
              – Lorenzo Quarisa
              Aug 29 at 13:54












            up vote
            2
            down vote










            up vote
            2
            down vote









            How about
            $$
            f(x)=|log x|?
            $$



            The above is a wrong answer, since it does not fulfill 3. So let's try again.



            If $0<a<1$, then
            $$
            int_0^1|log |x-a||,dx=1 - (1 - a) log(1 - a)- a log a<2.
            $$
            Order the rationals in $[0,1]$ in a sequence $a_n$, let $f_n(x)=bigl|,log |x-a_n|,bigr|$ and
            $$
            f(x)=sum_n=1^infty2^-nf_n(x) .
            $$
            Clearly $f$ is positive. Moreover
            $$
            int_0^1f(x),dx=sum_n=1^infty2^-nint_0^1f_n(x),dxle2 sum_n=1^infty2^-n<infty,
            $$
            so that $f$ is finite almost everywhere and in $L^1$. We have to check that $fin L^p$ for $p>1$. Since
            $$
            |f|_plesum_n=1^infty2^-n|f_n|_p,
            $$
            it is enough to show that $|f_n|_ple C_p$ for some constant $C_p$ depending only on $p$. The function $|x-a|^1/(2p)bigl|,log |x-a|bigr|$ is continuous, and in particular bounded, as a function of $(x,a)$ on $[0,1]times[0,1]$. It follows that there exists a constant $C_p$ such that
            $$
            bigl|,log |x-a|^pbigr|lefracC_psqrt,quad0le xle1,quad0le ale1,
            $$
            and the claim follows.






            share|cite|improve this answer














            How about
            $$
            f(x)=|log x|?
            $$



            The above is a wrong answer, since it does not fulfill 3. So let's try again.



            If $0<a<1$, then
            $$
            int_0^1|log |x-a||,dx=1 - (1 - a) log(1 - a)- a log a<2.
            $$
            Order the rationals in $[0,1]$ in a sequence $a_n$, let $f_n(x)=bigl|,log |x-a_n|,bigr|$ and
            $$
            f(x)=sum_n=1^infty2^-nf_n(x) .
            $$
            Clearly $f$ is positive. Moreover
            $$
            int_0^1f(x),dx=sum_n=1^infty2^-nint_0^1f_n(x),dxle2 sum_n=1^infty2^-n<infty,
            $$
            so that $f$ is finite almost everywhere and in $L^1$. We have to check that $fin L^p$ for $p>1$. Since
            $$
            |f|_plesum_n=1^infty2^-n|f_n|_p,
            $$
            it is enough to show that $|f_n|_ple C_p$ for some constant $C_p$ depending only on $p$. The function $|x-a|^1/(2p)bigl|,log |x-a|bigr|$ is continuous, and in particular bounded, as a function of $(x,a)$ on $[0,1]times[0,1]$. It follows that there exists a constant $C_p$ such that
            $$
            bigl|,log |x-a|^pbigr|lefracC_psqrt,quad0le xle1,quad0le ale1,
            $$
            and the claim follows.







            share|cite|improve this answer














            share|cite|improve this answer



            share|cite|improve this answer








            edited Aug 29 at 12:03

























            answered Aug 29 at 10:54









            Julián Aguirre

            65.2k23894




            65.2k23894











            • Oops. Did not read the "any interval" part.
              – Julián Aguirre
              Aug 29 at 11:20










            • I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
              – Lorenzo Quarisa
              Aug 29 at 12:09











            • Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
              – Lorenzo Quarisa
              Aug 29 at 12:26










            • @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
              – Julián Aguirre
              Aug 29 at 13:45











            • @JuliánAguirre Oh, that is true also. Thanks.
              – Lorenzo Quarisa
              Aug 29 at 13:54
















            • Oops. Did not read the "any interval" part.
              – Julián Aguirre
              Aug 29 at 11:20










            • I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
              – Lorenzo Quarisa
              Aug 29 at 12:09











            • Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
              – Lorenzo Quarisa
              Aug 29 at 12:26










            • @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
              – Julián Aguirre
              Aug 29 at 13:45











            • @JuliánAguirre Oh, that is true also. Thanks.
              – Lorenzo Quarisa
              Aug 29 at 13:54















            Oops. Did not read the "any interval" part.
            – Julián Aguirre
            Aug 29 at 11:20




            Oops. Did not read the "any interval" part.
            – Julián Aguirre
            Aug 29 at 11:20












            I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
            – Lorenzo Quarisa
            Aug 29 at 12:09





            I see that both our answers could be wrong, because we need $fin L^p$ for all $p>0$ and to prove $fin L^p$ we use Minkowski's inequality which only works when $pgeq 1$.
            – Lorenzo Quarisa
            Aug 29 at 12:09













            Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
            – Lorenzo Quarisa
            Aug 29 at 12:26




            Never mind, I see that the Minkowski inequality still holds for $pin (0,1)$ as long as we multiply by a coefficient depending only on $p$, so this exampe still works.
            – Lorenzo Quarisa
            Aug 29 at 12:26












            @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
            – Julián Aguirre
            Aug 29 at 13:45





            @LorenzoQuarisa Since $[0,1]$ has finite measure, $L^qsubset L^p$ if $qle p$.
            – Julián Aguirre
            Aug 29 at 13:45













            @JuliánAguirre Oh, that is true also. Thanks.
            – Lorenzo Quarisa
            Aug 29 at 13:54




            @JuliánAguirre Oh, that is true also. Thanks.
            – Lorenzo Quarisa
            Aug 29 at 13:54










            up vote
            1
            down vote













            Let
            $$mathbbQcap [0,1]=leftx_nright_nin mathbbN $$
            and
            $$f_n(x)=-log|x-x_n| $$
            so that $f_ngeq0$, $f_nin L^p(0,1)$ for all $pin (0,infty)$, and
            $$|f_n|_p^p=int_0^1|f_n|^p=int_0^1|log|x-x_n||^pdx leqint_-1^1|log |x||^pdx=:C<infty$$
            Let
            $$f=sum_nin mathbbNfrac12^nf_n $$
            then by Minkowski's inequality and Minkowski's inequality for $pin (0,1)$, we have
            $$|f|_pleq maxleft2^frac1p-1;1rightsum_nin mathbbNfrac12^n|f_n|_pleq max left2^frac1p-1;1rightC^1/psum_nin mathbbNfrac12^n<infty $$
            Hence $fin L^p$ for all $pin (0,infty)$, and for any interval $Isubset [0,1]$, there is $nin mathbbN$ such that $x_nin I$, and so $$operatornameess,supfgeq operatornameess,supf_n=+infty$$






            share|cite|improve this answer


























              up vote
              1
              down vote













              Let
              $$mathbbQcap [0,1]=leftx_nright_nin mathbbN $$
              and
              $$f_n(x)=-log|x-x_n| $$
              so that $f_ngeq0$, $f_nin L^p(0,1)$ for all $pin (0,infty)$, and
              $$|f_n|_p^p=int_0^1|f_n|^p=int_0^1|log|x-x_n||^pdx leqint_-1^1|log |x||^pdx=:C<infty$$
              Let
              $$f=sum_nin mathbbNfrac12^nf_n $$
              then by Minkowski's inequality and Minkowski's inequality for $pin (0,1)$, we have
              $$|f|_pleq maxleft2^frac1p-1;1rightsum_nin mathbbNfrac12^n|f_n|_pleq max left2^frac1p-1;1rightC^1/psum_nin mathbbNfrac12^n<infty $$
              Hence $fin L^p$ for all $pin (0,infty)$, and for any interval $Isubset [0,1]$, there is $nin mathbbN$ such that $x_nin I$, and so $$operatornameess,supfgeq operatornameess,supf_n=+infty$$






              share|cite|improve this answer
























                up vote
                1
                down vote










                up vote
                1
                down vote









                Let
                $$mathbbQcap [0,1]=leftx_nright_nin mathbbN $$
                and
                $$f_n(x)=-log|x-x_n| $$
                so that $f_ngeq0$, $f_nin L^p(0,1)$ for all $pin (0,infty)$, and
                $$|f_n|_p^p=int_0^1|f_n|^p=int_0^1|log|x-x_n||^pdx leqint_-1^1|log |x||^pdx=:C<infty$$
                Let
                $$f=sum_nin mathbbNfrac12^nf_n $$
                then by Minkowski's inequality and Minkowski's inequality for $pin (0,1)$, we have
                $$|f|_pleq maxleft2^frac1p-1;1rightsum_nin mathbbNfrac12^n|f_n|_pleq max left2^frac1p-1;1rightC^1/psum_nin mathbbNfrac12^n<infty $$
                Hence $fin L^p$ for all $pin (0,infty)$, and for any interval $Isubset [0,1]$, there is $nin mathbbN$ such that $x_nin I$, and so $$operatornameess,supfgeq operatornameess,supf_n=+infty$$






                share|cite|improve this answer














                Let
                $$mathbbQcap [0,1]=leftx_nright_nin mathbbN $$
                and
                $$f_n(x)=-log|x-x_n| $$
                so that $f_ngeq0$, $f_nin L^p(0,1)$ for all $pin (0,infty)$, and
                $$|f_n|_p^p=int_0^1|f_n|^p=int_0^1|log|x-x_n||^pdx leqint_-1^1|log |x||^pdx=:C<infty$$
                Let
                $$f=sum_nin mathbbNfrac12^nf_n $$
                then by Minkowski's inequality and Minkowski's inequality for $pin (0,1)$, we have
                $$|f|_pleq maxleft2^frac1p-1;1rightsum_nin mathbbNfrac12^n|f_n|_pleq max left2^frac1p-1;1rightC^1/psum_nin mathbbNfrac12^n<infty $$
                Hence $fin L^p$ for all $pin (0,infty)$, and for any interval $Isubset [0,1]$, there is $nin mathbbN$ such that $x_nin I$, and so $$operatornameess,supfgeq operatornameess,supf_n=+infty$$







                share|cite|improve this answer














                share|cite|improve this answer



                share|cite|improve this answer








                edited Aug 29 at 12:28

























                answered Aug 29 at 11:54









                Lorenzo Quarisa

                2,734315




                2,734315



























                     

                    draft saved


                    draft discarded















































                     


                    draft saved


                    draft discarded














                    StackExchange.ready(
                    function ()
                    StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f2898216%2fan-unbounded-function-with-integrability-for-all-p0%23new-answer', 'question_page');

                    );

                    Post as a guest













































































                    這個網誌中的熱門文章

                    How to combine Bézier curves to a surface?

                    Mutual Information Always Non-negative

                    Why am i infinitely getting the same tweet with the Twitter Search API?