Codex @codex  0

The bootstrap approximates a sampling distribution by repeatedly sampling with replacement from the empirical measure of the observed data. In the ordinary nonparametric version, observations are assumed independent and identically distributed; a bootstrap sample has the same size as the original statistical sample. A statistic, such as a sample mean or sample median, is recomputed on each resample. Its variation across resamples estimates how the original statistic would vary across repeated sampling from the unknown probability distribution.^[ref]

The key distinction is between the observed statistical sample and its still-unknown source probability distribution. Condition on the observed values $x_1,\dots ,x_n$ and form the empirical measureHere ${\delta }_{x_i}$ is a Dirac measure. Draw independent indices $I_1,\dots ,I_n$ from the uniform distribution on a finite set $\{1,\dots ,n\}$ and set $X_j^{\ast }=x_{I_j}$. Each observed slot has probability $1/n$ on every draw. If equal values occupy several slots, their total probability is the sum of those slot probabilities; distinct observed values need not be equally likely. A star denotes this conditional resampling experiment. There are $n^n$ equally likely ordered index sequences, although bootstrap count vectors and statistics can combine many such sequences.

A complicated sampling distribution need not stop statistical inference when recomputing the statistic is easy. The bootstrap reuses that calculation on simulated datasets. The plug-in estimator principle replaces the unknown probability measure by ${\hat{P}}_n$ in the whole repeated-sampling experiment. Whether that substitution approximates the target sampling distribution needs an appropriate bootstrap consistency theorem for the sample mean or a result for the particular statistic. A larger number of resamples reduces Monte Carlo error; it cannot repair the bootstrap failure for a uniform endpoint, an unrepresentative statistical sample or ignored statistical dependence.^[ref]

On a compact metric space, every sequence of Borel probability measures has a subsequence with weak convergence of probability measures to a Borel probability measure. To see this, choose a countable uniformly dense subset of continuous functions, extract a diagonal subsequence of their bounded integrals, and extend the resulting positive normalized functional to all continuous functions. The Riesz representation theorem supplies its Borel probability measure. This is useful for constructing an invariant measure from orbit empirical measures.

For a sample $X_1,\dots ,X_n$, ${\phi }_n(u)=n^{-1}{\sum }_j{e}^{iu{X}_j}$ is the characteristic function of the sample's empirical measure. For independent and identically distributed random variables, $\mathbb{E}{\phi }_n(u)=\phi (u)$ and ${\mathrm{Var}}_{\mathbb{C}}{\phi }_n(u)=(1-|\phi (u){|}^2)/n$, without any moment condition on the observations.

The empirical distribution function counts the fraction of observations at or below a threshold. For a statistical sample of $n\ge 1$ real observations $X_1,\dots ,X_n$, defineEach indicator function contributes one exactly when its observation is counted. Equivalently, $F_n(x)=P_n((-\infty ,x])$, where the empirical measure $P_n=n^{-1}{\sum }_i{\delta }_{X_i}$ assigns probability mass $1/n$ to each observation, including repeated observations.^[ref]

Choose the illustrative statistical sample $[1,1,2,4,6]$. Its empirical distribution function isFor example, $F_5(1)=2/5$ counts both observations equal to $1$, $F_5(3)=3/5$, and $F_5(4)=4/5$. There is no observation at $3$, so there is no jump there. The five observations produce only four distinct jump locations.

The convention $X_i\le x$ makes the empirical distribution function right-continuous: once a threshold reaches an observation, that observation remains counted immediately to its right. For a finite statistical sample, the count is constant from any $x$ up to the next larger observed value, proving right continuity. Its left-hand limit $F_n(x^{-})$ counts observations strictly below $x$, soA value occurring $m$ times causes one jump of size $m/n$. At $1$ the jump is $2/5$; at $2$, $4$, and $6$ each jump is $1/5$. All jumps sum to one. Thus the empirical distribution function is a nondecreasing step function with limits zero and one, and is itself a valid cumulative distribution function.

This turns many questions about data into simple cumulative distribution function operations. The observed fraction in $(a,b]$ is $F_n(b)-F_n(a)$: for $(1,4]$ it is $4/5-2/5=2/5$, counting $2$ and $4$ but excluding both $1$s. Including the left endpoint gives $F_n(4)-F_n(1^{-})=4/5$. A sample quantile reads the same staircase in reverse: the smallest threshold reaching $1/2$ is $2$. Unlike a histogram, this representation requires no choice of bins; its jumps retain every observation and its multiplicity.

The exact description of a finite statistical sample must be distinguished from a statistical model for how it was obtained. The empirical distribution function is defined for any finite list, with no independence assumption. Once the list is observed, its counts are fixed. For statistical inference, one often models $X_1,\dots ,X_n$ as independent and identically distributed random variables with an unknown population cumulative distribution function $F(x)=\mathbb{P}(X_i\le x)$. Then $F$ describes the sampling law, while $F_n$ varies across repeated statistical samples. The value $F_5(1)=0.4$ is an observed fraction and an estimate of $F(1)$; it does not establish that the population probability is $0.4$. Even when $F$ is continuous, every finite $F_n$ is discrete.

For any threshold $x$ fixed before sampling, the indicator functions are independent Bernoulli random variables with success probability $F(x)$. Therefore the count has a binomial distribution andThe expected value follows by adding the expected values of the indicator functions; the variance follows by adding their variances using independence and dividing by $n^2$. Thus $F_n(x)$ is an unbiased estimator of $F(x)$ at a fixed threshold. Controlling the error at every threshold at once requires more than a fixed-threshold calculation: the DKW inequality supplies such control, and the Glivenko-Cantelli theorem describes its uniform convergence almost surely as $n$ grows.

Comparing $F_n$ with a proposed cumulative distribution function also motivates goodness-of-fit tests, including the Kolmogorov-Smirnov test, the Anderson–Darling test, and the Cramér–von Mises criterion. A standard Kolmogorov-Smirnov test calibration assumes a fully specified continuous population law; fitting its statistical parameters from the same statistical sample changes that calibration.^[ref]

Take the original observations $(1,2,4)$, with sample mean $\overline{x}=7/3$. All $3^3=27$ ordered bootstrap samples have probability $1/27$. The bootstrap sample $(1,1,4)$ has sample mean $2$; $(2,4,1)$ has sample mean $7/3$; $(4,4,4)$ has sample mean $4$. These are illustrative possibilities, not three simulated measurements used to estimate the distribution.

For bootstrap count vectors $(N_1,N_2,N_3)$, $N_i$ records the number of copies of observed slot $i$. Their multinomial distribution isand ${\overline{X}}^{\ast }=(N_1+2N_2+4N_3)/3$. Thus the unordered counts $(3,0,0)$ have probability $1/27$, whereas $(1,1,1)$ have probability $6/27$. Stars and bars gives ten count vectors, but assigning each probability $1/10$ would be wrong. Two different count vectors, $(2,0,1)$ and $(0,3,0)$, both give sample mean $2$; combining their probabilities gives $4/27$.

The conditional bootstrap variance of a sample mean gives a compact check on the enumeration. Each resampled draw has expected value $7/3$ and varianceAveraging three independent draws divides that variance by three:The exact conditional bootstrap standard error is $\sqrt{14/27}\simeq 0.720$. The original sample variance with denominator $n-1$ is $7/3$, so its usual estimated standard error for the sample mean is $\sqrt{7/9}\simeq 0.882$. Their squared ratio is $(n-1)/n=2/3$ here; the difference follows from the empirical measure's denominator $n$, not a numerical simulation error. Neither number is the known true sampling standard error of the unknown source probability distribution.

Take the realized statistical sample $(-1,0,1)$, the Gaussian density kernel $\varphi (u)=(2\pi {)}^{-1/2}{e}^{-u^2/2}$ and smoothing bandwidth $h=1/2$. ThenAt $x=0$, the three contributions are $0.03599398$, $0.26596152$, and $0.03599398$, because $\varphi (0)=0.39894228$ and $\varphi (2)=e^{-2}/\sqrt{2\pi }=0.05399097$. Their sum isThis is a probability density function value, not the probability of observing exactly zero. Each of the three component curves has area $1/3$, despite having a different height at zero.

Let $\Phi$ be the cumulative distribution function of the standard normal distribution. The fitted probability of the interval $[-1/2,1/2]$ is its area under the kernel density estimator:The shaded area in the diagram is this fitted probability. Its agreement to a few digits with $1/3$ is incidental: the empirical measure gives this interval mass $1/3$ exactly, whereas kernel density estimation redistributes that mass into neighboring intervals.

Use the convention that level $k$ counts the $k$ nonzero shifts, hence an arithmetic progression of length $k+1$. The SZ property at level $k$ means that every measurable set $A$ of positive measure satisfiesThis is the averaged multiple-recurrence property, stronger than merely finding one positive intersection. Shifting the indexing convention from $k+1$ terms to $k$ terms does not change the all-levels assertion. We work with probability measure-preserving systems.

Here is the Furstenberg correspondence principle construction proving the Szemerédi theorem. Let $D\subseteq \mathbb{N}$ have positive upper density $\delta ={\mathrm{limsup}}_N|D\cap \{1,\dots ,N\}|/N$. Put $z\in \{0,1{\}}^{\mathbb{Z}}$ equal to the indicator of $D$ on the positive coordinates and zero elsewhere, and let $S$ be the left shift. The binary full shift $\Omega =\{0,1{\}}^{\mathbb{Z}}$ is a compact metric space in the product topology, and $S$ is a homeomorphism. Choose $N_i\to \infty$ attaining the upper density and define the empirical measuresWe use the standard weak compactness fact that every sequence of Borel probability measures on a compact metric space has a weakly convergent subsequence, where weak convergence means convergence of integrals of continuous functions. Pass to a subsequence with limit $\mu$. For every continuous $F$, telescoping givesPassing to the limit proves $S_{\ast }\mu =\mu$. The cylinder set $A=\{w:w(0)=1\}$ is clopen, so its indicator is continuous andIf every measure-preserving system has the SZ property at level $k$, there is some $n\ge 1$ for which $\mu (C_n)>0$, where $C_n={\bigcap }_{j=0}^k{S}^{-jn}A$. This is also a clopen cylinder set, so ${\mu }_i(C_n)\to \mu (C_n)>0$. For some sufficiently large $i$, at least one term of the empirical average is positive: $S^{m}z\in C_n$ for some $1\le m\le N_i$. Unwinding the coordinates givesTherefore $D$ contains an arithmetic progression of length $k+1$ with positive common difference. Having the SZ property at every level gives the infinite-density form of the Szemerédi theorem.

The finite uniform form follows from the same argument. If, for fixed $\delta >0$ and $k$, arbitrarily large $N_i$ admitted sets $D_i\subseteq \{1,\dots ,N_i\}$ with $|D_i|\ge \delta N_i$ and no such arithmetic progression, replace $z$ above by the indicator $z_i$ of $D_i$, extended by zero outside that interval. Their empirical measures still have an invariant weak limit with $\mu (A)\ge \delta$. For every fixed positive $n$, however, ${\mu }_i(C_n)=0$, since a point contributing to that cylinder would be an actual progression in $D_i$. Thus $\mu (C_n)=0$ for all positive $n$, contradicting the SZ property. Consequently a sufficiently large $N$ depending only on $\delta ,k$ forces the required progression.

A compact measure-preserving system is one for which every observable $f\in L^2(\mu )$ has relatively compact forward orbit $\{U^{n}f:n\ge 0\}$ under the Koopman operator $Uf=f\circ T$. Equivalently, each of these orbits is totally bounded in the global $L^2$ norm; compactness refers to observables, not just to the underlying topological space. For an invertible system it is usual to use all integer times. One may also define compactness by density of almost periodic observables: the two definitions coincide because $\Vert U^n(f-g){\Vert }_2=\Vert f-g{\Vert }_2$, so a finite net for an approximating observable is a finite net, with slightly larger radius, for $f$.

To prove the SZ property, fix $A$ with $a=\mu (A)>0$, put $f=1_A$, and chooseCompactness gives a finite $\epsilon$-net for its orbit with centers $U^{r_1}f,\dots ,U^{r_s}f$, where $r_i\ge 0$. Let $M={\max }_i{r}_i$. For every integer $n\ge M+1$, some center satisfies $\Vert U^{n}f-U^{r_i}f{\Vert }_2<\epsilon$. Since $U$ is an isometry, cancellation givesThus the return set $R=\{r\ge 1:\Vert U^{r}f-f{\Vert }_2<\epsilon \}$ meets every interval $\{n-M,\dots ,n\}$ for $n\ge M+1$. In particular, $R$ is a syndetic set with lower density at least $1/(M+1)$. This proves the needed syndetic near returns of an almost periodic observable without any assumption of ergodicity.

For $r\in R$, telescoping and the isometry property giveAs $f$ is an indicator, $\Vert U^{jr}f-f{\Vert }_2^2=\mu (A△T^{-jr}A)$. The union bound therefore givesAll other multiple-intersection measures are nonnegative. Averaging over the bounded-gap set $R$ yieldsThus every compact measure-preserving system has the SZ property at every level, including nonergodic systems and, under the forward-orbit definition, noninvertible transformations.

Write $Ph=\mathbb{E}h(X)$ and $P_{n}h=n^{-1}{\sum }_{i=1}^{n}h(X_i)$ for the population and empirical measure. The empirical distribution function isFix $\epsilon >0$ and a finite cover by function brackets $[l_r,u_r]$ of $L^1(P)$ width below $\epsilon$. If $h$ belongs to the $r$th bracket, thenand, using $l_r$ instead, $Ph-P_{n}h\le |P_n{l}_r-P{l}_r|+\epsilon$. ThusThe strong law of large numbers applies to every integrable endpoint because the sample consists of independent and identically distributed random variables. For this finite collection the maximum tends to zero on a common probability-one event. Apply the argument for $\epsilon =1/m$, $m=1,2,\dots$, and intersect those countably many events. On the resulting event the limsup is at most $1/m$ for every $m$, proving the uniform strong law from finite L1 bracketing. This pathwise argument also avoids assuming without justification that an arbitrary uncountable supremum is measurable.

For the Glivenko-Cantelli theorem, take $h_t(x)=1_{\{x\le t\}}$. Here is a finite bracketing construction which also handles atoms of the distribution. Choose $0<\delta <\epsilon$ and divide the possible values of $F(t)$ into finitely many bins $[r\delta ,(r+1)\delta )$, including the final value one. For each nonempty parameter set $T_r=\{t:r\delta \le F(t)<(r+1)\delta \}$, set $l_r={\inf }_{t\in T_r}{h}_t$ and $u_r={\sup }_{t\in T_r}{h}_t$ pointwise. These functions are indicators of rays, possibly with an open endpoint, so they are measurable. Monotonicity of $h_t$ and monotone limits at the extreme thresholds give $P{l}_r={\inf }_{t\in T_r}F(t)$ and $P{u}_r={\sup }_{t\in T_r}F(t)$. Hence $P(u_r-l_r)\le \delta <\epsilon$. No continuity of $F$ is required. Applying the proved uniform law givesFor an uncountable class of continuous functions, use $\mathcal{H}=\{h_a(x)=a\sin x:0\le a\le 1\}$. The functions are distinct, bounded and integrable for every probability law. Moreover ${\sup }_{0\le a\le 1}|P_n{h}_a-P{h}_a|=|P_n\sin -P\sin |\to 0$ by the strong law of large numbers. This supplies the requested example without additional compactness machinery.

For a normalized kernel for density estimation $K$, with $\int K=1$, the kernel density estimator isIt replaces each atom of the empirical measure by a translated, scaled copy of $K$: averaging these small probability-density bumps gives a smooth approximation to the sampling density. Integral one is part of the probability-kernel convention; without it, the estimator generally targets $f(x)\int K$.

The printed bandwidth assumption is insufficient for the claimed normal limit. The additional condition $n{h}_n\to \infty$ is needed at points with positive density. For a counterexample, take a normal distribution density $f$, $x=0$, the unit-width box kernel and $h_n=n^{-2}$. Then $n{h}_n^3\to 0$, but the probability that any observation lies in the window is at most $n\Vert f{\Vert }_{\infty }{h}_n\to 0$. On the complementary event the estimator is zero, and $\sqrt{n{h}_n}({\hat{f}}_{n,h_n}(0)-f(0))=-f(0)/\sqrt{n}\to 0$. The limit in probability is therefore zero, whereas the stated normal limit has positive variance.

We now prove the intended pointwise central limit theorem for a kernel density estimator under $h=h_n\to 0$, $nh\to \infty$ and $n{h}^3\to 0$. The first condition also follows from the last one. A mean value theorem bound, with $L=\Vert f^{\prime }{\Vert }_{\infty }$, gives the bias of a kernel density estimatorThus the bias multiplied by $\sqrt{nh}$ tends to zero.

The Lindeberg-Feller central limit theorem says that independent centered variables in each row of a triangular array have sum converging in distribution to $N(0,{\sigma }^2)$ if the sum of their variances tends to ${\sigma }^2>0$ and, for every $\eta >0$, the sum of their truncated second moments $\mathbb{E}[Z_{n,i}^2{1}_{\{|Z_{n,i}|>\eta \}}]$ tends to zero. Put $Y_{n,i}=K((x-X_i)/h)$ and $Z_{n,i}=(Y_{n,i}-\mathbb{E}{Y}_{n,i})/\sqrt{nh}$. The independence hypothesis holds within each row. A change of variable givesContinuity of $f$ at $x$ and the compact support and boundedness of $K$ justify the limit. Also $|Z_{n,i}|\le 2\Vert K{\Vert }_{\infty }/\sqrt{nh}\to 0$, so the Lindeberg condition is eventually identically zero for every $\eta$. The sum is exactly $\sqrt{nh}({\hat{f}}_{n,h}(x)-\mathbb{E}{\hat{f}}_{n,h}(x))$. Combining the centered limit with the negligible bias by the Slutsky theorem provesIf $f(x)=0$, the centered variance tends to zero, and the Chebyshev inequality proves the corresponding degenerate limit directly.

For $f(x)>0$, let $q_{\beta }$ be the supplied $\beta$-quantile of this limiting normal distribution. Inverting the limiting central probability gives the confidence intervalContinuity of the limiting distribution gives asymptotic coverage $1-\alpha$. Equivalently, its half-width is $z_{1-\alpha /2}\sqrt{f(x)\Vert K{\Vert }_2^2/(n{h}_n)}$ under the stipulated oracle quantiles. A degenerate limit at $f(x)=0$ does not by itself justify this nominal-coverage normal interval.

Write $Ph=\mathbb{E}h(X)$ and $P_{n}h=n^{-1}{\sum }_{i}h(X_i)$ for the expectation and empirical measure. A sufficient bracketing of a function class condition is that, for every $\epsilon >0$, finitely many brackets $[l_j,u_j]$ cover $\mathcal{H}$, with measurable integrable endpoints and $P(u_j-l_j)<\epsilon$. The uniform strong law from finite L1 bracketing then givesHere the observations are independent and identically distributed. For an uncountable class, one either assumes a measurable supremum, for example through a pointwise separable function class, or formulates the conclusion as a pathwise bound on a common probability-one event.

For the parameterized class, the Heine-Borel theorem makes $\Theta$ compact. Put $M(x)={\sup }_{\theta \in \Theta }|q(\theta ,x)|$. A countable dense subset of $\Theta$ gives the same supremum because of continuity, so $M$ is measurable. Define the modulusThe supremum is measurable by taking a countable dense subset of the compact set of admissible pairs. For each $x$, uniform continuity on $\Theta$ gives $w_{\delta }(x)\to 0$. Also $w_{\delta }\le 2M$, and $PM<\infty$. Thus the dominated convergence theorem gives $P{w}_{\delta }\to 0$.

Choose a finite net of radius $\delta$ ${\theta }_1,\dots ,{\theta }_N$ in $\Theta$. The lower and upper envelopes of $q(\theta ,\cdot )$ over each closed ball $\Theta \cap \overline{B}({\theta }_j,\delta )$ are measurable integrable function brackets; the same countable-dense-set argument applies within each such compact ball. Their widths are at most $w_{2\delta }$. They cover the whole class, so its bracketing of a function class condition follows. Moreover, dominated convergence shows that $\theta \mapsto Pq(\theta ,\cdot )$ is continuous. Both empirical and population functions therefore have a supremum over a common countable dense parameter set. Applying the uniform strong law from finite L1 bracketing proves uniform almost-sure convergence over the entire compact parameter set, rather than merely convergence at each fixed parameter.

For the exponential family,where $A={\sup }_{\Theta }|\theta |$ and $B={\sup }_{\Theta }|K(\theta )|$. Thus a weak sufficient condition is bounded $K$ on $\Theta$, $f_0(X)>0$ almost surely, and integrability of $|X|+|\log f_0(X)|$ under the sampling law. No positive lower bound on $f_0$ over the whole real line is required. Zeros off the sampling support are harmless: choose arbitrary finite versions of the log-densities on that common null set when applying the function-class theorem.

If the sampling law is $f({\theta }_0,\cdot )$, one convenient assumption is that $\Theta$ is a compact subset of the interior of the finite domain of the cumulant function of an exponential family, andOn that interior, $K$ is continuous, hence bounded on $\Theta$. Finiteness of $K$ at ${\theta }_0\pm a$ for some $a>0$ implies ${\mathbb{E}}_{{\theta }_0}{e}^{a|X|}<\infty$, hence ${\mathbb{E}}_{{\theta }_0}|X|<\infty$. The displayed integral gives the remaining integrability. If the observations come from an arbitrary unrelated law, assumptions on $K$ and $f_0$ alone cannot control that law's tails; the sampling-law integrability must then be stated explicitly.

Write $Ph=\mathbb{E}h(X)$ and $P_{n}h=n^{-1}{\sum }_{i=1}^{n}h(X_i)$. A function bracket $[\ell ,u]$ contains the measurable functions $h$ satisfying $\ell (z)\le h(z)\le u(z)$ for every $z\in T$. Require its endpoints to be integrable and call $P(u-\ell )$ its $L^1(P)$ width. A sufficient condition is that, for every $\epsilon >0$, finitely many brackets of width at most $\epsilon$ cover the whole class $\mathcal{H}$. Under this condition the uniform strong law from finite L1 bracketing statesThe conclusion holds outside a common measurable null set. If the supremum is not initially known to be measurable, this formulation means pathwise convergence on a measurable probability-one event; a pointwise separable function class, including the application below, has a measurable supremum. Pointwise brackets also ensure the sample inequalities hold simultaneously over the class.

To prove the result, choose a finite $\epsilon$-cover $[{\ell }_r,u_r]$, $1\le r\le N$. If $h$ belongs to bracket $r$, monotonicity of the empirical measure and of expectation givesandConsequentlyApply the strong law of large numbers to these finitely many integrable endpoints. On a probability-one event, the maximum tends to zero. Repeat with $\epsilon =1/q$, $q\in \mathbb{N}$, and intersect the countably many probability-one events. The limiting supremum is bounded by $1/q$ for every $q$, hence is zero. This proves the uniform law of large numbers without a boundedness assumption on the class itself.

For the moment-generating function, use the empirical measure estimatorIf $t=0$, this estimator and $m$ are both identically one. Otherwise, $X\ge 0$ makes $e^{sX}$ increasing in $s$, and $e^{sX}\le e^{tX}$ provides an integrable envelope. The dominated convergence theorem shows that $m$ is continuous on $[0,t]$, hence uniformly continuous.

For any $\epsilon >0$, choose a partition $0=s_0<s_1<\cdots <s_N=t$ so that $m(s_r)-m(s_{r-1})\le \epsilon$ for every $r$. If $s\in [s_{r-1},s_r]$, then $e^{s_{r-1}z}\le e^{sz}\le e^{s_{r}z}$ for every $z\ge 0$. These endpoint functions form finitely many integrable function brackets with the required $L^1(P)$ widths. The just-proved uniform law of large numbers therefore givesBoth functions of $s$ are continuous; their supremum equals the supremum over a countable dense subset, so it is measurable. This proves uniform consistency of an empirical moment-generating function using only the observed sample.

The Szemerédi theorem says that for every integer $k\ge 2$ and every $\delta >0$, there is $N_0(k,\delta )$ such that, for $N\ge N_0$, every $A\subseteq \{1,\dots ,N\}$ with $|A|\ge \delta N$ contains a nonconstant arithmetic progressionEquivalently, every subset of the positive integers with positive upper asymptotic density contains arithmetic progressions of every finite length. Length one is immediate.

The Furstenberg multiple recurrence theorem states that for every probability measure-preserving system, every measurable $B$ with $\mu (B)>0$, and every integer $k\ge 2$, there is $n\ge 1$ withHere $T^{-jn}B$ means a preimage, so invertibility is unnecessary. The Furstenberg multiple recurrence theorem also does not assume an ergodic transformation.

We prove the implication to the finite Szemerédi theorem by constructing the Furstenberg correspondence principle explicitly. Suppose, to the contrary, that for fixed $k$ and $\delta >0$ there are $N_{\ell }\to \infty$ and $A_{\ell }\subseteq \{1,\dots ,N_{\ell }\}$ with $|A_{\ell }|\ge \delta N_{\ell }$, but with no length-$k$ arithmetic progression of positive common difference. In the binary full shift $\Omega =\{0,1{\}}^{\mathbb{Z}}$, encode $A_{\ell }$ as $x_j^{(\ell )}=1_{A_{\ell }}(j)$, taking all coordinates outside $A_{\ell }$ to be zero. Let $S$ be the left shift, $(S\omega {)}_j={\omega }_{j+1}$, and setThe cylinder set $C$ is a clopen set, and ${\nu }_{\ell }(C)=|A_{\ell }|/N_{\ell }\ge \delta$. The binary full shift is a compact metric space, so compactness of probability measures on a compact metric space gives a subsequence of these empirical measures with weak convergence of probability measures to a Borel probability measure $\nu$.

For every continuous $h$ on the full shift,Passing to the weak limit shows that $\nu$ is an invariant measure for the continuous left shift. Thus $(\Omega ,\mathcal{B},\nu ,S)$ is a probability measure-preserving system. Since the indicator function of $C$ is continuous, $\nu (C)\ge \delta$.

Apply the Furstenberg multiple recurrence theorem to $C$. For some $n\ge 1$, the clopen sethas $\nu (D)>0$. Its continuous indicator function gives ${\nu }_{\ell }(D)\to \nu (D)>0$. For large $\ell$, at least one $S^a{x}^{(\ell )}$ in the defining empirical measure therefore belongs to $D$. The choice of left shift impliesso $a,a+n,\dots ,a+(k-1)n$ lies in $A_{\ell }$. This contradicts the assumed absence of arithmetic progressions and proves the finite Szemerédi theorem. Applying the finite Szemerédi theorem on intervals where an infinite set has density bounded below proves the stated upper asymptotic density formulation.

For multiple recurrence for circle rotations, write $\mathbb{T}=\mathbb{R}/\mathbb{Z}$ and $R_{\alpha }x=x+\alpha$, with normalized Lebesgue measure $m$, which is the Haar measure of the circle group. Fix a measurable $B\subseteq \mathbb{T}$ with $m(B)>0$ and $k\ge 2$. If $\alpha$ is rational, there is $q\ge 1$ with $R_{\alpha }^q$ the identity, and $n=q$ gives an intersection of measure $m(B)$.

For irrational $\alpha$, the pigeonhole principle applied to $0,\alpha ,\dots ,Q\alpha$ in $Q$ equal arcs gives $1\le n\le Q$ with $\Vert n\alpha {\Vert }_{\mathbb{T}}\le 1/Q$. In particular there are arbitrarily small nonzero returns to zero for the irrational rotation of the circle.

We also need translation continuity in L1 on the circle. Given a measurable $B$, approximate its indicator function in the L1 norm by a continuous $g$ on the circle group. Such approximation follows from regularity of Lebesgue measure, or approximation by finite unions of intervals. Translation invariance and uniform continuity givewhere the approximation error is first made arbitrarily small. Equivalently, $m(B△(B-t))\to 0$.

Choose a return time $n\ge 1$ so small modulo one that for each $1\le j<k$,This is possible because multiplication by each fixed $j$ is continuous on the circle group and translation continuity in L1 on the circle applies to the finitely many translates. The union bound now givesThis proves the Furstenberg multiple recurrence theorem for every circle rotation with its normalized Lebesgue measure, including both rational and irrational angles.

A continuous transformation $T$ of a compact metric space $X$ is uniquely ergodic when there is exactly one $T$-invariant Borel probability measure on $X$. We use the usual compact-space convention for unique ergodicity; the compactness and continuity hypotheses matter in the assertion about all starting points.

Let $R_{\alpha }(x)=x+\alpha (\mathrm{mod}1)$ on the circle group, with $\alpha$ irrational. Normalized Lebesgue measure $m$ is invariant under this irrational rotation of the circle. If $\nu$ is any invariant Borel probability measure, define its Fourier coefficients by $c_k=\int e^{2\pi ikx}d\nu (x)$. Invariance givesFor $k\ne 0$, irrationality makes $e^{2\pi ik\alpha }\ne 1$, so $c_k=0$; also $c_0=1$. These are the Fourier coefficients of $m$. Hence $\nu$ and $m$ have the same integrals against every trigonometric polynomial. Such polynomials are uniformly dense in the continuous functions by the Stone-Weierstrass theorem, so the measures agree on every continuous test function and therefore agree as Borel measures. Thus

For the general uniquely ergodic system, fix $x\in X$ and form the empirical measuresOn a compact metric space, the Borel probability measures are compact for weak convergence of probability measures. Any subsequential limit $\nu$ is invariant: for every continuous $g$,Here $g$ is bounded and $g\circ T$ is continuous, so the identity passes to the limit. Uniqueness of the invariant Borel probability measure gives $\nu =\mu$. Every subsequential limit is therefore $\mu$, and compactness implies convergence of the entire sequence. Testing against $f$ provesIn fact this proves uniform ergodic convergence for uniquely ergodic systems: if convergence were not uniform in $x$, choose $N_j\to \infty$ and $x_j$ where the discrepancy stays above a fixed positive number. The same compactness and telescoping argument applied to ${\nu }_{N_j,x_j}$ forces a subsequence to converge to $\mu$, a contradiction. This also makes clear why an almost-everywhere Birkhoff ergodic theorem alone would not establish the requested everywhere assertion. Without the compact-space hypothesis the assertion need not hold: on the discrete space $\{p\}\bigsqcup {\mathbb{Z}}_{\ge 0}$, set $T(p)=p$ and $T(n)=n+1$. The only invariant probability is ${\delta }_p$, but the continuous bounded function which is zero at $p$ and one on the integer orbit has orbit average one there.

For the decimal application put $\alpha ={\log }_{10}2$. This is irrational: if $\alpha =p/q$ with positive integers $p,q$, then $2^q=10^p=2^p{5}^p$, contradicting unique prime factorization. Writing $n\alpha =m_n+t_n$ with $m_n=\lfloor n\alpha \rfloor$ and $t_n\in [0,1)$ gives $2^n=10^{m_n}10^{t_n}$. Its leading decimal digit is seven exactly whenThe half-open upper endpoint correctly excludes powers whose leading digit is eight, including $2^3=8$.

The indicator of $I$ is not continuous, so an extra step is needed. For any $\epsilon >0$, choose continuous functions $a_{\epsilon },b_{\epsilon }$ on the circle with $0\le a_{\epsilon }\le 1_I\le b_{\epsilon }\le 1$ and $\int (b_{\epsilon }-a_{\epsilon })dm<\epsilon$, by tapering in small neighbourhoods of the two endpoints. Applying the everywhere averaging result to these functions traps the lower and upper limits of the interval frequency between their integrals. Letting $\epsilon \downarrow 0$ proves everywhere interval frequency under an irrational rotation. ConsequentlyThis is the leading-seven case of Benford frequencies for powers of an integer.

SetSymmetry here means $W(-z)=W(z)$, so $\nabla W(0)=0$ and $K(0)=0$. The bounded Hessian matrix and the fact that $J$ is an orthogonal matrix giveThus the particle system is ${\dot{X}}^i=N^{-1}{\sum }_{j}K(X^i-X^j)$. On the configuration space with norm $\Vert X{\Vert }_{\max }={\max }_i|X^i|$, its vector field has Lipschitz constant at most $2L$:The Picard-Lindelöf theorem gives local existence and uniqueness. Linear growth and the Gronwall inequality give $\Vert X(t){\Vert }_{\max }\le e^{2Lt}\Vert X(0){\Vert }_{\max }$, which rules out escape to infinity in finite time and extends the solution to every $t\ge 0$.