A new test for tail index with application to Danish fire loss data

The conditional approach of Hill's estimator depends on a threshold choice. This may give different results in a statistical test when different thresholds are used. Motivated by the uniformly most powerful test, this article proposes a new test for heavy tail of various degrees. Simulation shows that the test is applicable and its power is superior to three existing methods in the literature. An example of the Danish fire loss data elucidates the inconsistent conclusion in existing statistical tests and demonstrates the consistent conclusion the proposed test leads to. A new and affirmative insight into the data is that the variance does not exist.

[1]  Paul Deheuvels,et al.  Kernel Estimates of the Tail Index of a Distribution , 1985 .

[2]  R. Khan,et al.  Sequential Tests of Statistical Hypotheses. , 1972 .

[3]  Nicole A. Lazar,et al.  Statistics of Extremes: Theory and Applications , 2005, Technometrics.

[4]  A. McNeil.  Estimating the Tails of Loss Severity Distributions Using Extreme Value Theory , 1997, ASTIN Bulletin.

[5]  J. Beirlant,et al.  On univariate extreme value statistics and the estimation of reinsurance premiums , 2006 .

[6]  E. Haeusler,et al.  On Asymptotic Normality of Hill's Estimator for the Exponent of Regular Variation , 1985 .

[7]  Armelle Guillou,et al.  A diagnostic for selecting the threshold in extreme value analysis , 2001 .

[8]  M. Kulldorff,et al.  A Maximized Sequential Probability Ratio Test for Drug and Vaccine Safety Surveillance , 2011 .

[9]  J. Teugels,et al.  Statistics of Extremes , 2004 .

[10]  Robert H. Berk,et al.  LOCALLY MOST POWERFUL SEQUENTIAL TESTS , 1975 .

[11]  Sidney Resnick,et al.  Smoothing the Hill Estimator , 1997, Advances in Applied Probability.

[12]  S. Csörgo,et al.  Estimating the tail index , 1998 .

[13]  Peter Hall,et al.  Using the bootstrap to estimate mean squared error and select smoothing parameter in nonparametric problems , 1990 .

[14]  C. Klüppelberg,et al.  Modelling Extremal Events , 1997 .

[15]  B. M. Hill,et al.  A Simple General Approach to Inference About the Tail of a Distribution , 1975 .

[16]  D. Mason.  Laws of Large Numbers for Sums of Extreme Values , 1982 .

[17]  J. Andel.  Sequential Analysis , 2022, The SAGE Encyclopedia of Research Design.

[18]  Peter B. Gregory,et al.  A Sequential Clinical Trial for Testing p1 = p2. , 1980 .

[19]  R. Huisman,et al.  Tail-Index Estimates in Small Samples , 2001 .

[20]  Edgar Kaufmann,et al.  Selecting the optimal sample fraction in univariate extreme value estimation , 1998 .

[21]  Wai Keung Li,et al.  A threshold approach for peaks-over-threshold modeling using maximum product of spacings , 2010 .

[22]  J. Teugels,et al.  Tail Index Estimation, Pareto Quantile Plots, and Regression Diagnostics , 1996 .

[23]  A. McNeil,et al.  Estimation of tail-related risk measures for heteroscedastic financial time series: an extreme value approach , 2000 .

[24]  D. Siegmund.  Sequential Analysis: Tests and Confidence Intervals , 1985 .

[25]  Barbara Szyszkowicz,et al.  Asymptotic methods in probability and statistics : a volume in honour of Miklós Csörgö , 1998 .

[26]  J. Pickands.  Statistical Inference Using Extreme Order Statistics , 1975 .

[27]  L. Haan,et al.  Using a Bootstrap Method to Choose the Sample Fraction in Tail Index Estimation , 2000 .

[28]  Robert L. Wolpert,et al.  Statistical Inference , 2019, Encyclopedia of Social Network Analysis and Mining.