Laws of Small Numbers: Extremes and Rare Events

Laws of Small Numbers: Extremes and Rare Events

Michael Falk

Language: English

Pages: 509

ISBN: 3034800088

Format: PDF / Kindle (mobi) / ePub


Since the publication of the first edition of this seminar book in 1994, the theory and applications of extremes and rare events have enjoyed an enormous and still increasing interest. The intention of the book is to give a mathematically oriented development of the theory of rare events underlying various applications. This characteristic of the book was strengthened in the second edition by incorporating various new results. In this third edition, the dramatic change of focus of extreme value theory has been taken into account: from concentrating on maxima of observations it has shifted to large observations, defined as exceedances over high thresholds. One emphasis of the present third edition lies on multivariate generalized Pareto distributions, their representations, properties such as their peaks-over-threshold stability, simulation, testing and estimation. Reviews of the 2nd edition: "In brief, it is clear that this will surely be a valuable resource for anyone involved in, or seeking to master, the more mathematical features of this field" David Stirzaker, Bulletin of the London Mathematical Society "Laws of Small Numbers can be highly recommended to everyone who is looking for a smooth introduction to Poisson approximations in EVT and other fields of probability theory and statistics. In particular, it offers an interesting view on multivariate EVT and on EVT for non-iid observations, which is not presented in a similar way in any other textbook" Holger Drees, Metrika

Elementary Linear Algebra (7th Edition)

Lectures on Algebraic Geometry 1: Sheaves, Cohomology of Sheaves, and Applications to Riemann Surfaces (Aspects of Mathematics, Volume 35)

Six Degrees: The Science of a Connected Age

The Simpsons and Their Mathematical Secrets

The Grapes Of Math: How Life Reflects Numbers and Numbers Reflect Life

Applications of Lie Groups to Differential Equations (Graduate Texts in Mathematics, Volume 107)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

is, we consider only those observations which exceed the threshold t. The (t) process Nn is therefore called the point process of the exceedances. From Theorem 1.3.1 we know that we can write Nn(t) (·) = j≤Kt (n) (t) εV (t) +t (·), j (t) where the excesses V1 , V2 , . . . are independent replicates of a rv V (t) with df F (t) (·) := P (Z ≤ t + ·|Z ≥ t), and these are independent of the sample size Kt (n) := i≤n εZi ((t, ∞)). Without specific assumptions, the problem to determine F −1 (1 − q)

class of distributions possessing a regularly varying upper tail or, equivalently, with polynomially decreasing upper tails. The designation of super-heavy concerns right tails decreasing to zero at a slower rate, as logarithmic, for instance. This also means that the classical bible for inferring about rare events, the Extreme Value Theory, is no longer applicable, since we are in presence of distributions with slowly varying tails. We give a short overview of the peaks-over-threshold approach

(Yi,n ))i=1 , n (2.64) n and defining Q(i) n := Yn−i,n − Yn−k,n , q(Yn−k,n ) i = 0, 1, . . . , k − 1, (2.65) as well as Mn(1) := 1 k k−1 i=0 log(U (Yn−i,n )) − log(U (Yn−k,n )), (2.66) we get in turn Tn (k) =D = log(U (Yn,n )) − log(U (Yn−k,n )) k−1 i=0 log(U (Yn−i,n )) − log(U (Yn−k,n )) (0) = (2.67) (1) k Mn log(U (Yn,n )) − log(U (Yn−k,n )) log U Yn−k,n + Qk,n q(Yn−k,n ) k−1 log U Yn−k,n + i=0 (i) Qk,n q(Yn−k,n ) − log(U (Yn−k,n )) . (2.68) − log(U (Yn−k,n ))

estimation of general regression functionals T (F (· | x)) has been receiving increasing interest only some years ago (see, for example, Stute [424], H¨ ardle et al. [206], Samanta [404], Truong [447], Manteiga [317], Hendricks and Koenker [216], Goldstein and Messer [177]). Truncated Empirical Process Statistical inference based on (X1 , Y1 ), . . . , (Xn , Yn ) of a functional T (F (· | x)) has obviously to be based on those Yi among Y1 , . . . , Yn , whose corresponding Xi -values are close

1.1 Introduction The economist Ladislaus von Bortkiewicz, born 1868 in St. Petersburg (that Russian town, whose name was changed several times during this century: 1703-1914 St. Petersburg, 1914-1924 Petrograd, 1924-1991 Leningrad, since 1991 St. Petersburg again), Professor in Berlin from 1901 until his death in 1931, was presumably one of the first to recognize the practical importance of the Poisson approximation of binomial distributions. His book The law of small numbers [51] popularized

Download sample

Download