統計分布、統計学的分布

WordNet

the spatial or geographic property of being scattered about over a range, area, or volume; "worldwide in distribution"; "the distribution of nerve fibers"; "in complementary distribution" (同)dispersion
the act of distributing or spreading or apportioning
the commercial activity of transporting and selling goods from a producer to a consumer
(statistics) an arrangement of values of a variable showing their observed or theoretical frequency of occurrence (同)statistical_distribution
a datum that can be represented numerically
of or relating to statistics; "statistical population"

PrepTutorEJDIC

〈U〉(…への)『配分』,『分配』,配布,配達《+『to』+『名』》 / 〈U〉分布;分布区域 / 〈C〉配分されたもの,配給品 / 〈U〉(商品の)流通[組織,機構]
統計値,統計量
統計の,統計学上の

Wikipedia preview

出典(authority):フリー百科事典『ウィキペディア（Wikipedia）』「2018/10/01 10:27:00」(JST)

wiki en

The green curve, which asymptotically approaches heights of 0 and 1 without reaching them, is the true cumulative distribution function of the standard normal distribution. The grey hash marks represent the observations in a particular sample drawn from that distribution, and the horizontal steps of the blue step function (including the leftmost point in each step but not including the rightmost point) form the empirical distribution function of that sample. (Click here to load a new graph.)

In statistics, an empirical distribution function is the distribution function associated with the empirical measure of a sample. This cumulative distribution function is a step function that jumps up by $1/ n$ at each of the $n$ data points. Its value at any specified value of the measured variable is the fraction of observations of the measured variable that are less than or equal to the specified value.

The empirical distribution function is an estimate of the cumulative distribution function that generated the points in the sample. It converges with probability 1 to that underlying distribution, according to the Glivenko–Cantelli theorem. A number of results exist to quantify the rate of convergence of the empirical distribution function to the underlying cumulative distribution function.

Definition

Let $(x 1, \dots, x n)$ be independent, identically distributed real random variables with the common cumulative distribution function $F (t)$ . Then the empirical distribution function is defined as ^[1]^[2]

{\widehat {F}}_{n}(t)={\frac {{\mbox{number of elements in the sample}}\leq t}{n}}={\frac {1}{n}}\sum _{i=1}^{n}\mathbf {1} _{x_{i}\leq t},

where $\mathbf {1} _{A}$ is the indicator of event $A$ . For a fixed $t$ , the indicator $\mathbf {1} _{x_{i}\leq t}$ is a Bernoulli random variable with parameter $p = F (t)$ ; hence $n{\widehat {F}}_{n}(t)$ is a binomial random variable with mean $nF (t)$ and variance $nF (t)(1 - F (t))$ . This implies that ${\widehat {F}}_{n}(t)$ is an unbiased estimator for $F (t)$ .

However, in some textbooks, the definition is given as

${\widehat {F}}_{n}(t)={\frac {1}{n+1}}\sum _{i=1}^{n}\mathbf {1} _{x_{i}\leq t}$

 
   
     
       
         
           
             
               F
               ^
             
           
         
         
           n
         
       
       (
       t
       )
       =
       
         
           1
           
             n
             +
             1
           
         
       
       
         ∑
         
           i
           =
           1
         
         
           n
         
       
       
         
           1
         
         
           
             x
             
               i
             
           
           ≤
           t
         
       
     
   
   {\displaystyle {\widehat {F}}_{n}(t)={\frac {1}{n+1}}\sum _{i=1}^{n}\mathbf {1} _{x_{i}\leq t}}

^[3]^[4]

Small-sample properties

The mean of the empirical distribution is an unbiased estimator of the mean of the population distribution.

The variance of the empirical distribution times ${\tfrac {n}{n-1}}$

 
   
     
       
         
           
             n
             
               n
               −
               1
             
           
         
       
     
   
   {\displaystyle {\tfrac {n}{n-1}}}

is an unbiased estimator of the variance of the population distribution.

Asymptotic properties

Since the ratio $(n + 1)/ n$ approaches 1 as $n$ goes to infinity, the asymptotic properties of the two definitions that are given above are the same.

By the strong law of large numbers, the estimator $\scriptstyle {\widehat {F}}_{n}(t)$

 
   
     
       
         
           
             
               
                 F
                 ^
               
             
           
           
             n
           
         
         (
         t
         )
       
     
   
   {\displaystyle \scriptstyle {\widehat {F}}_{n}(t)}

converges to

F (t)

as

n \to \infty

almost surely, for every value of

t

:^[1]

{\widehat {F}}_{n}(t)\ {\xrightarrow {\text{a.s.}}}\ F(t);

thus the estimator $\scriptstyle {\widehat {F}}_{n}(t)$ is consistent. This expression asserts the pointwise convergence of the empirical distribution function to the true cumulative distribution function. There is a stronger result, called the Glivenko–Cantelli theorem, which states that the convergence in fact happens uniformly over $t$ :^[5]

\|{\widehat {F}}_{n}-F\|_{\infty }\equiv \sup _{t\in \mathbb {R} }{\big |}{\widehat {F}}_{n}(t)-F(t){\big |}\ {\xrightarrow {\text{a.s.}}}\ 0.

The sup-norm in this expression is called the Kolmogorov–Smirnov statistic for testing the goodness-of-fit between the empirical distribution $\scriptstyle {\widehat {F}}_{n}(t)$ and the assumed true cumulative distribution function $F$ . Other norm functions may be reasonably used here instead of the sup-norm. For example, the L²-norm gives rise to the Cramér–von Mises statistic.

The asymptotic distribution can be further characterized in several different ways. First, the central limit theorem states that pointwise, $\scriptstyle {\widehat {F}}_{n}(t)$

 
   
     
       
         
           
             
               
                 F
                 ^
               
             
           
           
             n
           
         
         (
         t
         )
       
     
   
   {\displaystyle \scriptstyle {\widehat {F}}_{n}(t)}

has asymptotically normal distribution with the standard

{\sqrt {n}}

 
   
     
       
         
           n
         
       
     
   
   {\displaystyle {\sqrt {n}}}

rate of convergence:^[1]

{\sqrt {n}}{\big (}{\widehat {F}}_{n}(t)-F(t){\big )}\ \ {\xrightarrow {d}}\ \ {\mathcal {N}}{\Big (}0,F(t){\big (}1-F(t){\big )}{\Big )}.

This result is extended by the Donsker’s theorem, which asserts that the empirical process $\scriptstyle {\sqrt {n}}({\widehat {F}}_{n}-F)$ , viewed as a function indexed by $\scriptstyle t\in \mathbb {R}$ , converges in distribution in the Skorokhod space $\scriptstyle D[-\infty ,+\infty ]$ to the mean-zero Gaussian process $\scriptstyle G_{F}=B\circ F$ , where $B$ is the standard Brownian bridge.^[5] The covariance structure of this Gaussian process is

\operatorname {E} [\,G_{F}(t_{1})G_{F}(t_{2})\,]=F(t_{1}\wedge t_{2})-F(t_{1})F(t_{2}).

The uniform rate of convergence in Donsker’s theorem can be quantified by the result known as the Hungarian embedding:^[6]

\limsup _{n\to \infty }{\frac {\sqrt {n}}{\ln ^{2}n}}{\big \|}{\sqrt {n}}({\widehat {F}}_{n}-F)-G_{F,n}{\big \|}_{\infty }<\infty ,\quad {\text{a.s.}}

Alternatively, the rate of convergence of $\scriptstyle {\sqrt {n}}({\widehat {F}}_{n}-F)$ can also be quantified in terms of the asymptotic behavior of the sup-norm of this expression. Number of results exist in this venue, for example the Dvoretzky–Kiefer–Wolfowitz inequality provides bound on the tail probabilities of $\scriptstyle {\sqrt {n}}\|{\widehat {F}}_{n}-F\|_{\infty }$ :^[6]

\Pr \!{\Big (}{\sqrt {n}}\|{\widehat {F}}_{n}-F\|_{\infty }>z{\Big )}\leq 2e^{-2z^{2}}.

In fact, Kolmogorov has shown that if the cumulative distribution function $F$ is continuous, then the expression $\scriptstyle {\sqrt {n}}\|{\widehat {F}}_{n}-F\|_{\infty }$ converges in distribution to $\scriptstyle \|B\|_{\infty }$ , which has the Kolmogorov distribution that does not depend on the form of $F$ .

Another result, which follows from the law of the iterated logarithm, is that ^[6]

\limsup _{n\to \infty }{\frac {{\sqrt {n}}\|{\widehat {F}}_{n}-F\|_{\infty }}{\sqrt {2\ln \ln n}}}\leq {\frac {1}{2}},\quad {\text{a.s.}}

and

\liminf _{n\to \infty }{\sqrt {2n\ln \ln n}}\|{\widehat {F}}_{n}-F\|_{\infty }={\frac {\pi }{2}},\quad {\text{a.s.}}

References

^ ^a ^b ^c van der Vaart, A.W. (1998). Asymptotic statistics. Cambridge University Press. p. 265. ISBN 0-521-78450-6.
^ PlanetMath Archived May 9, 2013, at the Wayback Machine.
^ Coles, S. (2001) An Introduction to Statistical Modeling of Extreme Values. Springer, p. 36, Definition 2.4. ISBN 978-1-4471-3675-0.
^ Madsen, H.O., Krenk, S., Lind, S.C. (2006) Methods of Structural Safety. Dover Publications. p. 148-149. ISBN 0486445976
^ ^a ^b van der Vaart, A.W. (1998). Asymptotic statistics. Cambridge University Press. p. 266. ISBN 0-521-78450-6.
^ ^a ^b ^c van der Vaart, A.W. (1998). Asymptotic statistics. Cambridge University Press. p. 268. ISBN 0-521-78450-6.

External links

UpToDate Contents

全文を閲覧するには購読必要です。 To read the full text you will need to subscribe.

1. 自動血液分析装置 automated hematology instrumentation
2. 小児の身体組成測定 measurement of body composition in children
3. Genetic contribution and pathophysiology of obesity
4. 血液透析における水の浄化システム water purification systems in hemodialysis
5. 血液透析における水質の維持 maintaining water quality for hemodialysis

English Journal

Anharmonic vibrational spectra and mode-mode couplings analysis of 2-aminopyridine.

Faizan M1, Alam MJ1, Afroz Z2, Bhat SA1, Ahmad S3.
Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.Spectrochim Acta A Mol Biomol Spectrosc.2018 Jan 5;188:26-31. doi: 10.1016/j.saa.2017.06.054. Epub 2017 Jul 4.
PMID 28689075

Comprehensive Intrametastatic Immune Quantification and Major Impact of Immunoscore on Survival.

Mlecnik B1, Van den Eynde M1, Bindea G1, Church SE1, Vasaturo A1, Fredriksen T1, Lafontaine L1, Haicheur N1, Marliot F1, Debetancourt D1, Pairet G1, Jouret-Mourin A1, Gigot JF1, Hubert C1, Danse E1, Dragean C1, Carrasco J1, Humblet Y1, Valge-Archer V1, Berger A1, Pagès F1, Machiels JP1, Galon J1.
Journal of the National Cancer Institute.J Natl Cancer Inst.2018 Jan 1;110(1). doi: 10.1093/jnci/djx123.
PMID 28922789

Understanding the spatial distribution of factors controlling topsoil organic carbon content in European soils.

Rial M1, Martínez Cortizas A1, Rodríguez-Lado L2.
The Science of the total environment.Sci Total Environ.2017 Dec 31;609:1411-1422. doi: 10.1016/j.scitotenv.2017.08.012. Epub 2017 Aug 6.
PMID 28797147

Japanese Journal

無線物理量への情報変換を用いた無線センサネットワーク

藤井威生,太田真衣,遠藤朋実
情報処理学会研究報告. UBI, [ユビキタスコンピューティングシステム] 2013-UBI-38(53), 1-6, 2013-05-09
センサネットワークの情報収集では,通常の無線通信と異なり,情報伝達の正確性よりも,多数のセンサノードから情報収集のリアルタイム化や低消費電力化が求められる場面がある.従来からの IEEE802.15.4/Zigbee などパケット交換に基づくセンサ情報の収集は,多数のノードが存在した場合の情報収集時に,パケット間の相互干渉を回避するために必要とされるプロトコルのオーバヘッドや,消費電力の増加が大き …
NAID 110009579769

Group-Separation Effect in Cell-Size Distribution of Origami Crease Patterns

Yamamoto Ken,Yamazaki Yoshihiro
J Phys Soc Jpn 82(4), 044803-044803-7, 2013-04-15
… Some statistical properties of origami (the art of paper folding) are investigated. …
NAID 150000107061

Investigation of Positive and Negative Bias Temperature Instability of High-\kappa Dielectric Metal Gate Metal--Oxide--Semiconductor-Field-Effect-Transistors by Random Telegraph Signal

Huang Da-Cheng,Gong Jeng,Huang Chih-Fang
Jpn J Appl Phys 52(3), 036503-036503-4, 2013-03-25
… It is shown that I_{\text{d-RTS}} degradation (\Delta I_{\text{d-RTS}}/I_{\text{d-RTS}}) in NBTI devices exhibits a wider amplitude distribution than the PBTI one. … As a result, the statistical analysis indicates that \Delta I_{\text{D,sat}}/I_{\text{D,sat}} of the p-MOSFET is larger than \Delta I_{\text{D,sat}}/I_{\text{D,sat}} of the n-MOSFET. …
NAID 150000106571