The effect of rounding off in samples

Dit artikel bevat een elementaire behandeling van afrondingseffecten in steekproeven.
Zoals men verwacht, neemt de variantie van het steekproefgermddelde door af ronding der waarnemingen toe met een factor 1 + b ²/12σ _x ², waarbij b het afrondings- interval en σ _x ² de stamvariantie is, in analogie met de bekende correcties van Sheppard. Verder wordt het effect van afronding op de variantie van de steek-proefvariantie onderzocht. Er wordt op gewezen, dat de afrondingsfout ε in één steekproef gecorreleerd is met de afwijking van de nauwkeurige — d.w.z. niet afgeronde — stochastische variabele x van zijn gemiddelde μ, hoewel deze correlatie gemiddeld nul is. Het gevolg is, dat de variantie van de steekproefvariantie toeneemt. Bij "gewone" waarschijnlijkheidsverdelingen, die continue naar nul gaan bij x →, is deze toename gelijk aan
1/ n (1/3 b ²σ _x ²+ 1/180 b ⁴)
waarbij n de grootte van de steekproef is. Bij normale verdelingen is het effect ongeveer twee maal zo groot als de toename volgens de Sheppard's correctie van de afgeronde variantie.     

Approximation for the cumulative and inverse gamma distribution

The gamma distribution function can be expressed in terms of the Normal distribution and density functions with sufficient accuracy for most practical purposes.
The distribution function for the density x^Λ-1e^-x/μ/μ^ΛΓ(A) on 0 -R(Λ){(1 + 1/1 2Λ) φ(z) + 11 -z/4Λ^1/2+2(z²+ 2)/45Λ] φ(z) /3 Λ^1/2} where φ(z)≅1/[1 +_e^{-2z(√2/π+z2 /28)}] and φ(z) = e^{-z2 /2}/√2π are the Normal distribution and density functions, y is the appropriate root of y-y²/6+y³/36-y⁴/270= In (x/Λμ), z= Λ^1/2 y, and R( Λ) is the remainder term in Stirling's approximation for In Γ(Λ).     

Uitbijternegerende schatters bij duplobepalingen

Outlyer-ignoring estimators for measurement in duplo.
By hypothesis a measurement u is the sum of two independent random variables, the normal random variable with expectation μ, and standard error σ, and a random error φ:

Basically two independent measurements u₁ and u₂ over u are to give the estimate x=1/2(u₁+ u₂) over μ.
However, to reduce the effect of the error φ on a final estimate of μ, one adds, according to a common practice, a third or even a fourth measurement u₃, u₄, in the case that the basic pair differs by more than a number A. For this extended set of measurements two outlyer-ignoring estimator y and z of μ are defined, and investigated against three specifications fo the error φ. Also an outlyer-ignoring estimate of σ is considered, and its application is illustrated by an example.     

Superpositie van twee frequentieverdelingen

Summary  (Superposition of two frequency distributions)
Notation:
n: number of observations
M: arithmetic mean
: standard deviation
μ_r: r^th moment coefficient
β₁: coefficient of skewness
β₂: coefficient of kurtosis.
The suffixes a and b apply to the component distributions. The suffix t applies to the resulting distributions.

The problem: Given the first r moments of two frequency distributions (to begin with μ₀). Find the first r moments of the distribution resulting from superposition of the two components ( r ≥ 5 ).
Formulae [1]. … [ 5 ] (§ 3 ) give the results in their most general form up to μ₄.
Some special cases are treated in § 4, and eight different cases of superposition of two normal distributions in § 5.
In § 6 some remarks are made about the reverse situation, i.e. the splitting into two normal components of a combined frequency distribution.     

The best quadratic estimator of the residual variance in regression analysis

De beste kwadratische schattingsfunctie van de storingsvariantie in regressie-analyse.
Dit artikel handelt over de schatting van de variantie σ² van de storingen in de regressieanalyse onder klassieke veronderstellingen: niet-stochastische waarden aangenomen door de verklarende variabelen en normaliteit, onafhankelijkheid en homoskedasticiteit van de storingen. Bekend is dat de schatting volgens maximale aannemelijkheid neerkomt op net bepalen van de kwadratensom van de volgens kleinste-kwadraten geschatte storingen en deling door T (het aantal waarne-mingen); voorts, dat de schatting die minimale variantie heeft binnen de klasse van schattingsfuncties die zuiver zijn en kwadratisch in de afhankelijke variabele (de beste zuivere kwadratische schattingsfunctie) gevonden wordt door genoemde kwadratensom te delen door T–A, waarbij λ het aantal te schatten coëfficiënten is [d.w.z. het aantal verklarende variabelen (+ 1 indien een constante term aanwezig is)]. Hier wordt aangetoond, dat de schattingsfunctie van σ² die een minimaal tweede moment heeft binnen de klasse van schattingsfuncties die kwadratisch zijn in de afhankelijke variabele (de beste kwadratische schattingsfunctie) gevonden wordt door de kwadratensom van de volgens kleinste kwadraten geschatte storingen te delen door T–Λ+ 2.     

A different solution of a problem posed by Theil and Schweitzer

De vraag naar de nauwkeurigste (d.i. kleinste tweede moment) schatter van de restvariantie a² bij lineaire regressieanalyse met isomore onafhankelijke normaal verdeelde residuen wordt opnieuw beantwoord, maar longs vrijwel geheel andere weg dan door THEIL en SCHWEITZER [3] bewandeld. Het blijkt bovendien dat deling van de residuele som van kwadraten door het bijbehorende aantal vrijheids-graden + 2 niet alleen in de klasse der kwadratische funk ties der waarnemingen maar zelfs in de klasse van alle veelvouden van zuivere schattingen van σ² de nauwkeurigste is. Gebruik wordt gemaakt van de in een appendix bewezen stelling, dat de traditionele schatter van σ², waarbij de residuele som van kwadraten door het bijbehorend aantal vrijheidsgraden wordt gedeeld, de nauwkeurigste is onder alle zuivere schatters.     

De omvang van de steekproef bij een enkelvoudig steekproefsysteem

Summary  (Sample size for a single sampling scheme).
The operating characteristic of a sampling scheme may be specified by the producers 1 in 20 risk point ( p ₁), at which the probability of rejecting a batch is 0.05, and the consumers 1 in 20 risk point ( p ₂) at which the probability of accepting a batch of that quality is also 0.05.
A nomogram is given (fig. 2) to determine for single sampling schemes and for given values of p₁ and p ₂ the necessary sample size ( n ) and the allowable number of defectives in the sample ( c ).
The nomogram may reversedly be used to determine the producers and consumers 1 in 20 risk points for a given single sampling scheme.
The curves in this nomogram were computed from a table of percentage points of the χ² distribution. For v > 30 Wilson and Hilferty's approximation to the χ² distribution was used.     

An improved empirical Bayes test for positive exponential families

We exhibit an empirical Bayes test δ^* _n for a decision problem using a linear error loss in a class of positive exponential families. This empirical Bayes test δ^* _n possesses the asymptotic optimality, and its associated regret converges to zero with rate n ⁻¹(ln n )⁶ This rate of convergence improves the previous results in the literature in the sense that a faster rate of convergence is achieved under much weaker conditions. Examples are presented to illustrate the performance of the empirical Bayes test δ^* _n     

Notes on the Markowitz portfolio selection method

A proof of the validity of Markowitz's critical line method is given for a more general situation than discussed by Markowitz. Next for the Markowitz case with a positive definite covariance matrix explicit expressions are derived for all efficient portfolios. Using these expressions it can be shown that the critical line in the (μ,α²) plane is a representation of a function which is not necessarily differentiable everywhere.     

PSEUDO-R2 MEASURES FOR SOME COMMON LIMITED DEPENDENT VARIABLE MODELS

Abstract. A large number of different Pseudo- R ² measures for some common limited dependent variable models are surveyed. Measures include those based solely on the maximized likelihoods with and without the restriction that slope coefficients are zero, those which require further calculations based on parameter estimates of the coefficients and variances and those that are based solely on whether the qualitative predictions of the model are correct or not. The theme of the survey is that while there is no obvious criterion for choosing which Pseudo- R ² to use, if the estimation is in the context of an underlying latent dependent variable model, a case can be made for basing the choice on the strength of the numerical relationship to the OLS- R ² in the latent dependent variable. As such an OLS- R ² can be known in a Monte Carlo simulation, we summarize Monte Carlo results for some important latent dependent variable models (binary probit, ordinal probit and Tobit) and find that a Pseudo- R ² measure due to McKelvey and Zavoina scores consistently well under our criterion. We also very briefly discuss Pseudo- R ² measures for count data, for duration models and for prediction-realization tables.     

On a crossroad of resampling plans: bootstrapping elementary symmetric polynomials

We investigate the validity of the bootstrap method for the elementary symmetric polynomials S ^{( k )} _n =( ⁿ _k )⁻¹Σ_{1≤ i 1}< ... < i _k ≤ n X _{i 1} ... X _{i k} of i.i.d. random variables X ₁, ..., X _n . For both fixed and increasing order k , as n→∞ the cases where μ=E X ₁[moe2]0, the nondegenerate case, and where μ=E X ₁=0, the degenerate case, are considered.     

Fourier inversion and the Hausdorff distance

This paper continues research done by F.H. Ruymgaart and the author. For a function f on R ^d we consider its Fourier transform F f and the functions f_M (M>0) derived from F f by the formula f_M(x) =( F( ε_M · F f ))(− x );, where the ε_M are suitable integrable functions tending to 1 pointwise as M →∞. It was shown earlier that, relative to a metric d _H , analogous to the Hausdorff distance between closed sets, one has d _H (f_M, f) = O( M ^−½) for all f in a certain class. We now show that, for such f , the estimate O( M ^−½) is optimal if and only if f has a discontinuity point.     

Selection from Logistic populations

Assume k ( k ≥ 2) independent populations π₁, π₂μ_k are given. The associated independent random variables X_i,( i = 1,2,… k ) are Logistically distributed with unknown means μ₁, μ₂, μ_k and equal variances. The goal is to select that population which has the largest mean. The procedure is to select that population which yielded the maximal sample value. Let μ₍₁₎≤μ₍₂₎≤…≤μ_(k) denote the ordered means. The probability of correct selection has been determined for the Least Favourable Configuration μ₍₁₎=μ₍₂₎==μ_{(k – 1)}=μ_(k)–δ where δ > 0. An exact formula for the probability of correct selection is given.     

First-Order Integer-Valued Autoregressive (INAR (1)) Process: Distributional and Regression Properties

Some properties of a first-order integer-valued autoregressive process (INAR)) are investigated. The approach begins with discussing the self-decomposability and unimodality of the 1-dimensional marginals of the process {X_n} generated according to the scheme X_n=α° X _n-i +e_n, where α° X _n-1 denotes a sum of X_{n - 1}, independent 0 - 1 random variables Y^(n-1), independent of X _n-1 with Pr -( y ^{(n - 1)}= 1) = 1 - Pr ( y ^(n-i)= 0) =α. The distribution of the innovation process ( e _n) is obtained when the marginal distribution of the process ( X _n) is geometric. Regression behavior of the INAR(1) process shows that the linear regression property in the backward direction is true only for the Poisson INAR(1) process.     

Minimax character of the two-sample χ2-test

Abstract  We consider the problem of testing equality of the mean vectors of two multivariate normal populations when covariates are present and the covariance matrix is known. As an application of the H unt -S tein theorem it is shown that the χ²-^-test of level a maximizes, among all level a tests, the minimum power on each of the contours where the χ²-^-test has constant power. A corollary is that the χ²-^-test is most stringent level a .     

Maxima voor de eenzijdige overschrijdingskans en voor de grootte van de gemiddelde overschrijding bij discontinue en continue, eentoppige verdelingen

The well-known inequality of Bienaymé-Tschébyschef (for short B-T), generalized by Camp and Meidell (for short C-M) for continuous, unimodal distributions gives specific limits for total probabilities outside the ± to limits.
In many cases however, especially in the field of industrial applications we are interested only in the probability of one tail of the distribution, which of course must be smaller than the limits given by the B-T and C-M formula.
For these cases the maximum probability of surpassing the to limit on one side equals under B-T conditions and under C-M conditions instead of the two-sided values of 1/t² and 47/9 · 1/t² respectively (cf e.g. Uspensky: "Introduction to mathematical probability", 1937, p. 198) These results set upper limits for the value of

Alternatively we may also set an upper limit for the integral

which measures in terms of σ the average amount by which the limit + tσ is exceeded. This problem is also discussed and under C-M conditions an upper limit

is derived.
Some practical applications of these results are considered.     

Het verdelen van steekproeven over subpopulaties bij accountantscontroles

Summary  "Stratificationprocedures for a typical auditing problem".
During the past ten years, much experience was gained in The Netherlands in using random sampling methods for typical auditing problems. Especially, a method suggested by VAN. HEERDEN [2] turned out to be very fruitful. In this method a register of entries is considered to be a population of T guilders, if all entries total up to T guilders. The sample size n ₀ is determined in such a way that the probability β not to find any mistake in the sample, if a fraction p ₀ or more of T is incorrect, is smaller than a preassigned value β₀. So n ₀ should satisfy (l- p )ⁿ⁰≤β₀ for p ≥ p ₀. A complication arises if it is not possible to postpone sampling until the whole population T is available. One then wants to take samples from a population which is growing up to T . Suppose one is going to take samples n _i from e.g. r subpopulations

Using the minimax procedure, it is shown, that in this case one should choose the sizes n _i equal to ( T _i/ T ) n ₀. The minimax-value of the probability not to find any incorrect guilder in the r samples, taken together is equal to β₀.     

Statistics for the contact process

A d -dimensional contact process is a simplified model for the spread of an infection on the lattice Z ^d . At any given time t ≥ 0 , certain sites x ∈ Z ^d are infected while the remaining once are healthy. Infected sites recover at constant rate 1, while healthy sites are infected at a rate proportional to the number of infected neighboring sites. The model is parametrized by the proportionality constant λ. If λ is sufficiently small, infection dies out (subcritical process), whereas if λ is sufficiently large infection tends to be permanent (supercritical process).
In this paper we study the estimation problem for the parameter λ of the supercritical contact process starting with a single infected site at the origin. Based on an observation of this process at a single time t , we obtain an estimator for the parameter λ which is consistent and asymptotically normal as t →∞     

Skewed Normal Variance-Mean Models for Asset Pricing and the Method of Moments

Financial returns (log-increments) data, Y _t , t = 1,2,…, are treated as a stationary process, with the common distribution at each time point being not necessarily symmetric.
We consider as possible models for the common distribution four instances of the General Normal Variance-Mean Model (GNVM), which is described by Y | V ∼ N ( a ( b + V ), c ²V + d² ) where V is a nonnegative random variable and a, b, c and d are constants. When V is Gamma distributed and d = 0, Y has the skewed Variance-Gamma distribution (VG). When V follows a Half Normal distribution and c = 0, Y has the well-known Skew Normal (SN) distribution. We also consider two cases where V is Exponentially distributed. Bounds for skewness and kurtosis in each case are found in terms of the moments of the V . These are useful in determining whether the Method of Moments for a given model is feasible. The problem of overdetermination of parameters via estimating equations is examined. 5 data sets of actual returns data, chosen because of their earlier occurrence in the literature, are analysed using each of the 4 models.     

On the generalized Poisson distribution

We use Euler's difference lemma to prove that, for θ > 0 and 0 ≤λ < 1, the function P _n defined on the non-negative integers by
P _n (θ, λ) = [θ(θ + n λ) ^{n −1}/ n !]e^{− n λ−θ}
defines a probability distribution, known as the Generalized Poisson Distribution.