Composite goodness-of-fit tests for left-truncated loss samples

Anna Chernobay, Svetlozar T. Rachev, Frank J. Fabozzi

Research output: Chapter in Book/Report/Conference proceedingChapter

19 Citations (Scopus)

Abstract

In many financial models, such as those addressing value at risk and ruin probabilities, the accuracy of the fitted loss distribution in the upper tail of the loss data is crucial. In such situations, it is important to test the fitted loss distribution for the goodness of fit in the upper quantiles, while giving lesser importance to the fit in the low quantiles and the center of the distribution of the data. Additionally, in many loss models the recorded data are left truncated with the number of missing data unknown. We address this gap in literature by proposing appropriate goodness-of-fit tests. We derive the exact formulae for several goodness-of-fit statistics that should be applied to loss models with left-truncated data where the fit of a distribution in the right tail of the dstribution is of central importance. We apply the proposed tests to real financial losses, using a variety of distributions fitted to operational loss and the natural catastrophe insurance claims data, which are subject to the recording thresholds of $1 and $25 million, respectively.

Original languageEnglish (US)
Title of host publicationHandbook of Financial Econometrics and Statistics
PublisherSpringer New York
Pages575-596
Number of pages22
ISBN (Electronic)9781461477501
ISBN (Print)9781461477495
DOIs
StatePublished - Jan 1 2015

Fingerprint

Goodness of Fit Test
Composite
Goodness of fit
Quantile
Tail
Truncated Data
Ruin Probability
Value at Risk
Catastrophe
Goodness of fit test
Missing Data
Insurance
Model
Statistics
Unknown
Loss distribution

ASJC Scopus subject areas

  • Economics, Econometrics and Finance(all)
  • Business, Management and Accounting(all)
  • Mathematics(all)

Cite this

Chernobay, A., Rachev, S. T., & Fabozzi, F. J. (2015). Composite goodness-of-fit tests for left-truncated loss samples. In Handbook of Financial Econometrics and Statistics (pp. 575-596). Springer New York. https://doi.org/10.1007/978-1-4614-7750-1_20

Composite goodness-of-fit tests for left-truncated loss samples. / Chernobay, Anna; Rachev, Svetlozar T.; Fabozzi, Frank J.

Handbook of Financial Econometrics and Statistics. Springer New York, 2015. p. 575-596.

Research output: Chapter in Book/Report/Conference proceedingChapter

Chernobay, A, Rachev, ST & Fabozzi, FJ 2015, Composite goodness-of-fit tests for left-truncated loss samples. in Handbook of Financial Econometrics and Statistics. Springer New York, pp. 575-596. https://doi.org/10.1007/978-1-4614-7750-1_20
Chernobay A, Rachev ST, Fabozzi FJ. Composite goodness-of-fit tests for left-truncated loss samples. In Handbook of Financial Econometrics and Statistics. Springer New York. 2015. p. 575-596 https://doi.org/10.1007/978-1-4614-7750-1_20
Chernobay, Anna ; Rachev, Svetlozar T. ; Fabozzi, Frank J. / Composite goodness-of-fit tests for left-truncated loss samples. Handbook of Financial Econometrics and Statistics. Springer New York, 2015. pp. 575-596
@inbook{3020d5ab006f4aabb17aea623766dee3,
title = "Composite goodness-of-fit tests for left-truncated loss samples",
abstract = "In many financial models, such as those addressing value at risk and ruin probabilities, the accuracy of the fitted loss distribution in the upper tail of the loss data is crucial. In such situations, it is important to test the fitted loss distribution for the goodness of fit in the upper quantiles, while giving lesser importance to the fit in the low quantiles and the center of the distribution of the data. Additionally, in many loss models the recorded data are left truncated with the number of missing data unknown. We address this gap in literature by proposing appropriate goodness-of-fit tests. We derive the exact formulae for several goodness-of-fit statistics that should be applied to loss models with left-truncated data where the fit of a distribution in the right tail of the dstribution is of central importance. We apply the proposed tests to real financial losses, using a variety of distributions fitted to operational loss and the natural catastrophe insurance claims data, which are subject to the recording thresholds of $1 and $25 million, respectively.",
author = "Anna Chernobay and Rachev, {Svetlozar T.} and Fabozzi, {Frank J.}",
year = "2015",
month = "1",
day = "1",
doi = "10.1007/978-1-4614-7750-1_20",
language = "English (US)",
isbn = "9781461477495",
pages = "575--596",
booktitle = "Handbook of Financial Econometrics and Statistics",
publisher = "Springer New York",
address = "United States",

}

TY - CHAP

T1 - Composite goodness-of-fit tests for left-truncated loss samples

AU - Chernobay, Anna

AU - Rachev, Svetlozar T.

AU - Fabozzi, Frank J.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - In many financial models, such as those addressing value at risk and ruin probabilities, the accuracy of the fitted loss distribution in the upper tail of the loss data is crucial. In such situations, it is important to test the fitted loss distribution for the goodness of fit in the upper quantiles, while giving lesser importance to the fit in the low quantiles and the center of the distribution of the data. Additionally, in many loss models the recorded data are left truncated with the number of missing data unknown. We address this gap in literature by proposing appropriate goodness-of-fit tests. We derive the exact formulae for several goodness-of-fit statistics that should be applied to loss models with left-truncated data where the fit of a distribution in the right tail of the dstribution is of central importance. We apply the proposed tests to real financial losses, using a variety of distributions fitted to operational loss and the natural catastrophe insurance claims data, which are subject to the recording thresholds of $1 and $25 million, respectively.

AB - In many financial models, such as those addressing value at risk and ruin probabilities, the accuracy of the fitted loss distribution in the upper tail of the loss data is crucial. In such situations, it is important to test the fitted loss distribution for the goodness of fit in the upper quantiles, while giving lesser importance to the fit in the low quantiles and the center of the distribution of the data. Additionally, in many loss models the recorded data are left truncated with the number of missing data unknown. We address this gap in literature by proposing appropriate goodness-of-fit tests. We derive the exact formulae for several goodness-of-fit statistics that should be applied to loss models with left-truncated data where the fit of a distribution in the right tail of the dstribution is of central importance. We apply the proposed tests to real financial losses, using a variety of distributions fitted to operational loss and the natural catastrophe insurance claims data, which are subject to the recording thresholds of $1 and $25 million, respectively.

UR - http://www.scopus.com/inward/record.url?scp=84939470724&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84939470724&partnerID=8YFLogxK

U2 - 10.1007/978-1-4614-7750-1_20

DO - 10.1007/978-1-4614-7750-1_20

M3 - Chapter

AN - SCOPUS:84939470724

SN - 9781461477495

SP - 575

EP - 596

BT - Handbook of Financial Econometrics and Statistics

PB - Springer New York

ER -