Empirical distribution of sorted Gaussian numbers












1














I wrote a small program that does the following :




  1. Pick $N$ independent standard Gaussian numbers (expected value : 0, standard deviation : 1). Call that list $L={y_1, ldots, y_N}$.

  2. Sort that list in increasing order : $tilde{L}=mathrm{sort}(L)$.

  3. Plot that list on the $[-1,1]$ interval using a regularly distributed grid $x_i=-1+frac{2i}{N-1}$, with $i=0,ldots,N-1$.


I found that the plot was similar to that of the inverse error function, only differing by a multiplicative factor $a>0$.



Inverse error function & sorted Gaussian numbers plot, with N=10^5



I made a linear regression to find an approximate value of $1.42104$ for $a$. The two functions are very close for $N=10^5$ :
Fitted inverse error function and sorted Gaussian numbers for N=10^5



I have two questions :




  1. What is the exact value of $a$ ?

  2. How to prove that the limit function is indeed $a*mathrm{inverf}$ as $Nto infty$ ?






probability normal-distribution probability-limit-theorems sorting







share|cite|improve this question




















  • 1




    What about $sqrt 2$ ?
    – Claude Leibovici
    13 hours ago
















1














I wrote a small program that does the following :




  1. Pick $N$ independent standard Gaussian numbers (expected value : 0, standard deviation : 1). Call that list $L={y_1, ldots, y_N}$.

  2. Sort that list in increasing order : $tilde{L}=mathrm{sort}(L)$.

  3. Plot that list on the $[-1,1]$ interval using a regularly distributed grid $x_i=-1+frac{2i}{N-1}$, with $i=0,ldots,N-1$.


I found that the plot was similar to that of the inverse error function, only differing by a multiplicative factor $a>0$.



Inverse error function & sorted Gaussian numbers plot, with N=10^5



I made a linear regression to find an approximate value of $1.42104$ for $a$. The two functions are very close for $N=10^5$ :
Fitted inverse error function and sorted Gaussian numbers for N=10^5



I have two questions :




  1. What is the exact value of $a$ ?

  2. How to prove that the limit function is indeed $a*mathrm{inverf}$ as $Nto infty$ ?






probability normal-distribution probability-limit-theorems sorting







share|cite|improve this question




















  • 1




    What about $sqrt 2$ ?
    – Claude Leibovici
    13 hours ago














1












1








1







I wrote a small program that does the following :




  1. Pick $N$ independent standard Gaussian numbers (expected value : 0, standard deviation : 1). Call that list $L={y_1, ldots, y_N}$.

  2. Sort that list in increasing order : $tilde{L}=mathrm{sort}(L)$.

  3. Plot that list on the $[-1,1]$ interval using a regularly distributed grid $x_i=-1+frac{2i}{N-1}$, with $i=0,ldots,N-1$.


I found that the plot was similar to that of the inverse error function, only differing by a multiplicative factor $a>0$.



Inverse error function & sorted Gaussian numbers plot, with N=10^5



I made a linear regression to find an approximate value of $1.42104$ for $a$. The two functions are very close for $N=10^5$ :
Fitted inverse error function and sorted Gaussian numbers for N=10^5



I have two questions :




  1. What is the exact value of $a$ ?

  2. How to prove that the limit function is indeed $a*mathrm{inverf}$ as $Nto infty$ ?






probability normal-distribution probability-limit-theorems sorting







share|cite|improve this question















I wrote a small program that does the following :




  1. Pick $N$ independent standard Gaussian numbers (expected value : 0, standard deviation : 1). Call that list $L={y_1, ldots, y_N}$.

  2. Sort that list in increasing order : $tilde{L}=mathrm{sort}(L)$.

  3. Plot that list on the $[-1,1]$ interval using a regularly distributed grid $x_i=-1+frac{2i}{N-1}$, with $i=0,ldots,N-1$.


I found that the plot was similar to that of the inverse error function, only differing by a multiplicative factor $a>0$.



Inverse error function & sorted Gaussian numbers plot, with N=10^5



I made a linear regression to find an approximate value of $1.42104$ for $a$. The two functions are very close for $N=10^5$ :
Fitted inverse error function and sorted Gaussian numbers for N=10^5



I have two questions :




  1. What is the exact value of $a$ ?

  2. How to prove that the limit function is indeed $a*mathrm{inverf}$ as $Nto infty$ ?






probability normal-distribution probability-limit-theorems sorting




probability normal-distribution probability-limit-theorems sorting






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited yesterday

























asked yesterday









Florian Omnès

266




266








  • 1




    What about $sqrt 2$ ?
    – Claude Leibovici
    13 hours ago














  • 1




    What about $sqrt 2$ ?
    – Claude Leibovici
    13 hours ago








1




1




What about $sqrt 2$ ?
– Claude Leibovici
13 hours ago




What about $sqrt 2$ ?
– Claude Leibovici
13 hours ago










1 Answer
1






active

oldest

votes


















1














What you observe is the convergence of the empirical distribution function to cumulative distribution function of the sample - more accurately, of the empirical quantiles to theoretical quantiles (= values of inverse cumulative distribution function).



Specifically, for a continuous increasing cdf, theoretical quantiles are given by
$$
x_q = F^{-1}(q) = sup{xinmathbb R: F(x)<q}, qin(0,1).
$$
The definition of empirical quantiles varies. For a sample $X_1,dots,X_n$ of iid variables they can e.g. be defined by
$$
hat x_q = X_{(lfloor nqrfloor +1)}, qin (0,1),
$$
where $X_{(1)}le dotsle X_{(n)}$ is the sorted sample. It is known that whenever the cdf $F$ is strictly increasing, $hat x_q to x_q$ for all $qin (0,1)$ with probability $1$ as the sample size $ntoinfty$.



In your case, $F(x) = Phi(x)$ is the standart normal cdf, which is related to the error function by $$Phi(x) = frac{1+mathrm{Erf}(x/sqrt{2})}{2},$$
so
$$
Phi^{-1}(y) = sqrt{2}operatorname{Erf}^{-1}(2y-1), yin (0,1).
$$
What you are doing is applying a similar transformation to the empirical quantiles, so the convergence to $sqrt{2}operatorname{Erf}^{-1} approx 1.4142 operatorname{Erf}^{-1}$ is not surprising.






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',
    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%2f3060431%2fempirical-distribution-of-sorted-gaussian-numbers%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









    1














    What you observe is the convergence of the empirical distribution function to cumulative distribution function of the sample - more accurately, of the empirical quantiles to theoretical quantiles (= values of inverse cumulative distribution function).



    Specifically, for a continuous increasing cdf, theoretical quantiles are given by
    $$
    x_q = F^{-1}(q) = sup{xinmathbb R: F(x)<q}, qin(0,1).
    $$
    The definition of empirical quantiles varies. For a sample $X_1,dots,X_n$ of iid variables they can e.g. be defined by
    $$
    hat x_q = X_{(lfloor nqrfloor +1)}, qin (0,1),
    $$
    where $X_{(1)}le dotsle X_{(n)}$ is the sorted sample. It is known that whenever the cdf $F$ is strictly increasing, $hat x_q to x_q$ for all $qin (0,1)$ with probability $1$ as the sample size $ntoinfty$.



    In your case, $F(x) = Phi(x)$ is the standart normal cdf, which is related to the error function by $$Phi(x) = frac{1+mathrm{Erf}(x/sqrt{2})}{2},$$
    so
    $$
    Phi^{-1}(y) = sqrt{2}operatorname{Erf}^{-1}(2y-1), yin (0,1).
    $$
    What you are doing is applying a similar transformation to the empirical quantiles, so the convergence to $sqrt{2}operatorname{Erf}^{-1} approx 1.4142 operatorname{Erf}^{-1}$ is not surprising.






    share|cite|improve this answer


























      1














      What you observe is the convergence of the empirical distribution function to cumulative distribution function of the sample - more accurately, of the empirical quantiles to theoretical quantiles (= values of inverse cumulative distribution function).



      Specifically, for a continuous increasing cdf, theoretical quantiles are given by
      $$
      x_q = F^{-1}(q) = sup{xinmathbb R: F(x)<q}, qin(0,1).
      $$
      The definition of empirical quantiles varies. For a sample $X_1,dots,X_n$ of iid variables they can e.g. be defined by
      $$
      hat x_q = X_{(lfloor nqrfloor +1)}, qin (0,1),
      $$
      where $X_{(1)}le dotsle X_{(n)}$ is the sorted sample. It is known that whenever the cdf $F$ is strictly increasing, $hat x_q to x_q$ for all $qin (0,1)$ with probability $1$ as the sample size $ntoinfty$.



      In your case, $F(x) = Phi(x)$ is the standart normal cdf, which is related to the error function by $$Phi(x) = frac{1+mathrm{Erf}(x/sqrt{2})}{2},$$
      so
      $$
      Phi^{-1}(y) = sqrt{2}operatorname{Erf}^{-1}(2y-1), yin (0,1).
      $$
      What you are doing is applying a similar transformation to the empirical quantiles, so the convergence to $sqrt{2}operatorname{Erf}^{-1} approx 1.4142 operatorname{Erf}^{-1}$ is not surprising.






      share|cite|improve this answer
























        1












        1








        1






        What you observe is the convergence of the empirical distribution function to cumulative distribution function of the sample - more accurately, of the empirical quantiles to theoretical quantiles (= values of inverse cumulative distribution function).



        Specifically, for a continuous increasing cdf, theoretical quantiles are given by
        $$
        x_q = F^{-1}(q) = sup{xinmathbb R: F(x)<q}, qin(0,1).
        $$
        The definition of empirical quantiles varies. For a sample $X_1,dots,X_n$ of iid variables they can e.g. be defined by
        $$
        hat x_q = X_{(lfloor nqrfloor +1)}, qin (0,1),
        $$
        where $X_{(1)}le dotsle X_{(n)}$ is the sorted sample. It is known that whenever the cdf $F$ is strictly increasing, $hat x_q to x_q$ for all $qin (0,1)$ with probability $1$ as the sample size $ntoinfty$.



        In your case, $F(x) = Phi(x)$ is the standart normal cdf, which is related to the error function by $$Phi(x) = frac{1+mathrm{Erf}(x/sqrt{2})}{2},$$
        so
        $$
        Phi^{-1}(y) = sqrt{2}operatorname{Erf}^{-1}(2y-1), yin (0,1).
        $$
        What you are doing is applying a similar transformation to the empirical quantiles, so the convergence to $sqrt{2}operatorname{Erf}^{-1} approx 1.4142 operatorname{Erf}^{-1}$ is not surprising.






        share|cite|improve this answer












        What you observe is the convergence of the empirical distribution function to cumulative distribution function of the sample - more accurately, of the empirical quantiles to theoretical quantiles (= values of inverse cumulative distribution function).



        Specifically, for a continuous increasing cdf, theoretical quantiles are given by
        $$
        x_q = F^{-1}(q) = sup{xinmathbb R: F(x)<q}, qin(0,1).
        $$
        The definition of empirical quantiles varies. For a sample $X_1,dots,X_n$ of iid variables they can e.g. be defined by
        $$
        hat x_q = X_{(lfloor nqrfloor +1)}, qin (0,1),
        $$
        where $X_{(1)}le dotsle X_{(n)}$ is the sorted sample. It is known that whenever the cdf $F$ is strictly increasing, $hat x_q to x_q$ for all $qin (0,1)$ with probability $1$ as the sample size $ntoinfty$.



        In your case, $F(x) = Phi(x)$ is the standart normal cdf, which is related to the error function by $$Phi(x) = frac{1+mathrm{Erf}(x/sqrt{2})}{2},$$
        so
        $$
        Phi^{-1}(y) = sqrt{2}operatorname{Erf}^{-1}(2y-1), yin (0,1).
        $$
        What you are doing is applying a similar transformation to the empirical quantiles, so the convergence to $sqrt{2}operatorname{Erf}^{-1} approx 1.4142 operatorname{Erf}^{-1}$ is not surprising.







        share|cite|improve this answer












        share|cite|improve this answer



        share|cite|improve this answer










        answered 13 hours ago









        zhoraster

        15.7k21752




        15.7k21752






























            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.





            Some of your past answers have not been well-received, and you're in danger of being blocked from answering.


            Please pay close attention to the following guidance:


            • 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.




            draft saved


            draft discarded














            StackExchange.ready(
            function () {
            StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3060431%2fempirical-distribution-of-sorted-gaussian-numbers%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

            Mario Kart Wii

            Image
            Please consider supporting The Cutting Room Floor on Patreon. Thanks for all your support! Mario Kart Wii From The Cutting Room Floor Jump to: navigation, search Mario Kart Wii Developer : Nintendo Publisher : Nintendo Platform : Wii Released in JP : April 10, 2008 Released in US : April 27, 2008 Released in EU : April 11, 2008 Released in AU : April 24, 2008 Released in KR : April 30, 2009 This game has unused animations. This game has unused areas. This game has unused code. This game has unused objects. This game has unused modes / minigames. This game has unused graphics. This game has unused models. This game has unused items. This game has unused music. This game has regional differences. This game has a notes page This game has a prerelease article Mario Kart Wii is the sixth game in the series (not counting Arcade editions), which intro...
            Read more

            The Binding of Isaac: Rebirth/Afterbirth

            Image
            Please consider supporting The Cutting Room Floor on Patreon. Thanks for all your support! The Binding of Isaac: Rebirth/Afterbirth From The Cutting Room Floor < The Binding of Isaac: Rebirth Jump to: navigation, search This is a sub-page of The Binding of Isaac: Rebirth . Afterbirth is the first expansion pack for Rebirth and features a variety of improvements. Contents 1 Unused Graphics 1.1 Old Man Keeper 1.2 Skinless Hush 1.3 Miscellaneous Unused Hush Sprites 1.4 Generosity achievement 1.5 Afterbirth+ Room Editor 1.6 Bearded Ultra Greed 1.7 Early Ultra Greed 1.8 Unused Brownie Expression 1.9 Frail's Spinning Head 1.10 Mom's Dead Hand Alt 2 Unused Audio 2.1 Unused Dead Womb Layer 2.2 Unused Hush Battle Outro 2.3 The Stain Grunt Unused Graphics Old Man Keeper Image file: gfxuimain me...
            Read more

            What does “Dominus providebit” mean?

            Image
            5 Does "Dominus providebit" mean "The Lord will provide" or "The Lord provides"? I once had a bit of an argument with a guy who studied Latin over that. It's an inscription that appears on the rim of some old Swiss coins that I have. meaning latin-to-english-translation grammar-identification share | improve this question edited Jan 18 at 15:27 V2Blast 105 4 asked Jan 17 at 22:49 Donald K. Donald K. 26 1 ...
            Read more