A review of methods for combining internal and external data

A fuzzy credibility model to estimate the Operational Value at Risk using internal and external data of risk events

“…These quantitative models combine internal and external data to overcome the limitation of modelling only internal risk event data only, which refers to the volume of data available, because extreme OpR events rarely occur. Thus, internal data is mixed with external data so that tail events can be modelled [20,21].…”

Operational Risk Management in Financial Institutions: A Literature Review

“…Berg-Yuen and Medova [54], based on a sample of 50 internationally operating banks for the period from 2005 to 2006, analyse the relationship between economic capital and required regulatory capital set aside for operational risk. The last study in this subgroup, conducted by Galloppo and Previati [6], introduces several approaches that mix internal and external data to estimate the frequency and severity of operational losses.…”

“…Since internal databases of single banks are generally biased towards high-frequency and low-severity events, as they lack a sufficient number of tail events that would limit accurate modelling of the tail part of loss distribution, Basel II requires banks using AMA to supplement their internal data with external data to measure operational risk capital [1]. However, as noted throughout the current article, the literature highlights the shortage of systematically recorded operational loss databases for conducting empirical research or testing the validity of newly developed models on "real" 15 operational loss data (see, e.g., [5][6][7][8]87]). Taking these claims into the consideration, we identify operational risk databases used in the assessed articles and provide an overview of the content as well as on implication's frequency of identified databases that are listed in Table 3.…”

“…To calculate operational risk capital, financial institutions are required to use four data elements: internal loss data, external loss data, scenario analysis and business environmental and internal control factors [1]. However, it is often suggested that operational risk data are scarce or inaccessible [5][6][7][8]. Considering this view, the present study attempts to provide empirical evidence to support these arguments.…”

“…For more details, see https://scholar.google.com/intl/us/scholar/ help.html#coverage (accessed on 07 October 2016). 6 Lagner and Knyphausen-Aufseß [4] collect articles from journals ranked with A and B by the German Academic Association of Business Research (VBA) and papers available in the Social Science Research Network (SSRN) over the last three years. For more detailed critique of journal ranking see [12].…”