3502(d) the institution shall use PD estimation techniques only with supporting analysis. The institution shall recognise the importance of judgemental considerations in combining results of techniques and in making adjustments for limitations of techniques and information;
3504(e) to the extent that an institution uses data on internal default experience for the estimation of PDs, the estimates shall be reflective of underwriting standards and of any differences in the rating system that generated the data and the current rating system. Where underwriting standards or rating systems have changed, the institution shall add a greater margin of conservatism in its estimate of PD;
3506(f) to the extent that the institution associates or maps its internal grades to the scale used by an ECAI or a similar organisation and then attributes the default rate observed for the external organisation’s grades to the institution’s grades, mappings shall be based on a comparison of internal rating criteria to the criteria used by the external organisation and on a comparison of the internal and external ratings of any common obligors. Biases or inconsistencies in the mapping approach or underlying data shall be avoided. The criteria of the external organisation underlying the data used for quantification shall be oriented to default risk only and not reflect transaction characteristics. The analysis undertaken by the institution shall include a comparison of the default definitions used, subject to the requirements in Article 178. The institution shall document the basis for the mapping;
3508(g) the institution may, to the extent that it uses statistical default prediction models, estimate PDs as the count weighted average of default-probability estimates for individual obligors in a given grade. The institution’s use of default probability models for this purpose shall meet the standards specified in Article 174;
3510(h) irrespective of whether an institution is using external, internal, or pooled data sources, or a combination of the three, for its PD estimation, the length of the underlying historical observation period used shall be at least five years for at least one source. If the available observation period spans a longer period for any source, and this data is relevant, this longer period shall be used. The data shall include a representative mix of good and bad years from the economic cycle relevant for the type of exposures.
3518(a) the institution shall estimate PDs by obligor grade, facility grade or pool from long run averages of one-year default rates over a representative mix of good and bad economic periods;
3522(c) the institution shall regard internal data for assigning exposures to grades or pools as the primary source of information for estimating loss characteristics. The institution may use external data (including pooled data) or statistical models for quantification provided that the following strong links both exist:
3528(d) if the institution derives long run average estimates of PD and LGD for retail exposures from an estimate of total losses and an appropriate estimate of PD or LGD, the process for estimating total losses shall meet the overall standards for estimation of PD and LGD set out in this Part, and the outcome shall be consistent with the concept of LGD as set out in point (a) of Article 181(1);
3530(e) irrespective of whether the institution is using external, internal or pooled data sources or a combination of the three, for its estimation of loss characteristics, the length of the underlying historical observation period used shall be at least five years for at least one source. If the available observations span a longer period for any source, and these data are relevant, this longer period shall be used. The data shall include a representative mix of good and bad years from the economic cycle relevant for the type of exposures;
3534An institution may, for purchased retail receivables, use external and internal reference data. The institution shall use all relevant data sources as points of comparison.
3550An institution shall, in quantifying the risk parameters to be associated with rating grades or pools, apply the following requirements specific to LGD estimates:
3552(a) the institution shall estimate LGDs by facility grade or pool on the basis of the average realised LGDs by facility grade or pool using all observed defaults within the data sources (default weighted average);
3555(i) the institution shall, subject to point (b)(ii), use LGD estimates that are appropriate for an economic downturn if those are more conservative than the long-run average;
3559(1) analyse the difference between the distribution of exposures over facility grades or pools, or over appropriate intervals in case of continuous facility scales, of the current portfolio before and during the downturn period; and
3561(2) if a substantial difference in the distribution of exposures is identified as a result of the analysis in point (b)(ii)(1), apply non-negative adjustments to its downturn LGD estimates in point (b)(i) to limit the impact of an economic downturn on risk-weighted exposure amounts;
3563(c) the institution shall consider the extent of any interdependence between the risk of the obligor and that of the collateral or collateral provider. Cases where there is a significant degree of dependence shall be addressed in a conservative manner;
3565(d) currency mismatches between the underlying obligation and the collateral shall be treated conservatively in the institution’s assessment of LGD;
3567(e) where LGD estimates take into account the existence of collateral under the LGD Modelling Collateral Method (but where the institution is not applying the approach set out in Article 169B), these estimates shall not solely be based on the collateral’s estimated market value. LGD estimates shall take into account the effect of the potential inability of the institution to expeditiously gain control of the collateral and liquidate it;
3574(i) the institution shall, subject to point (h)(ii), for the specific case of exposures already in default, ensure that the LGD in default reflects downturn conditions where the estimates of LGD in default that are appropriate for an economic downturn are more conservative than the long-run average LGD for defaulted exposures;
3576(ii) for the purpose of point (h)(i), the LGD in default should be increased above the level referred to in point (h)(i) where this is necessary to ensure that, for each exposure, the difference between the LGD estimate and BEEL given current economic circumstances and exposure status covers the institution’s estimate of the increase in loss rate caused by possible additional unexpected losses during the recovery period (i.e. between the date of default and the final liquidation of the exposure);
3578(i) to the extent that unpaid late fees have been capitalised in the institution’s income statement, they shall be added to the institution’s measure of exposure and loss;
3580(j) for exposures to corporates, estimates of LGD shall be based on data over a minimum of five years, increasing by one year each year after implementation until a minimum of seven years is reached, for at least one data source. If the available observation period spans a longer period for any source, and the data is relevant, this longer period shall be used.