Quantitative Credit Analyst (senior)

By | Careers, General, News | No Comments

Quantitative Credit Analyst (senior)

OSIS™ is looking for a full time quantitative credit analyst based at its office in The Hague. The analyst will work in an innovative environment of a dynamic analytics boutique. The position is full time and suited for a motivated individual with demonstrable affinity with financial industry. Candidates should have strong quantitative and communication skills with at least 3 years of relevant work experiences in academics or in the financial industry.

OSIS™ offers an attractive reward package, an investment in further education and an inspiring, international working environment with clients in Europe, North America, Asia and Australia. We work for tier 1 banks, insurance companies, asset managers and pension funds.

Responsibilities of the Quantitative Credit Analyst are:

  • Maintenance and improvement of the OSIS™ data infrastructure and suite of statistical models
  • Analysing asset-backed securities and loan portfolios under various stress scenarios
  • Producing credit analysis and research with macro-economic elements


  • Enthusiastic and creative team player
  • Strong quantitative analytical skills
  • English both verbal and in writing (one or two other European languages is a plus)
  • Experience with financial programming (preferably R and Shiny)
  • Demonstrable affinity with the financial industry

Applications package: CV and motivation
For sending your application please use recruitment@os-is.com and for more information please call us at  +31 20 70 85 769



Loan to Value driving the risk weights for residential mortgages

By | Bank Regulatory Capital vs Economic Capital, Regulations | No Comments

Basel 3 Standardised floor and default observations on residential mortgages in the EU

On 3 January 2017 the Basel Committee of Banking Supervision (BCBS) announced that it needs more work before the Committee can reach agreement on the package of proposals which would lead to Basel IV. One of the proposals was to introduce a Standardised floor. With the Standardised floor, the Committee wants to limit unexplained differences in the Risk Weighted Average (RWA) calculations of individual banks. Banks who have applied for the Internal Ratings Based Approach (IRBA) can calculate their own Probability of Default (PD), Loss Given Default (LGD) and Exposure at Default (EAD) estimates which are the inputs for the Basel RWA calculations. With the introduction of Standardised floor, the banks need to use the higher of RWA (determined through a new developed Standardised approach multiplied by a still-to-be determined percentage) and the RWA (based on their IRBA calculation).

Regarding the Standardised Approach for residential mortgages, the Committee have proposed to make the RWA calculation dependent on the value of the property in relation to the size of the loan, so-called loan-to-value (LtV). The higher the ratio – house value divided by loan value – the higher the RWA, thus the more the bank needs to put aside for regulatory capital.

OSIS have analysed the default performance of c. 10 million European residential mortgage loans published by the European DataWarehouse and compared the observations per LtV bucket between the various countries.

The observations show large discrepancies between two groups of countries. In one group we have Italy, Spain, Portugal and Ireland and in the other Sweden, Germany, The Netherlands, Belgium, France and the UK. On average the default rates in the first group are three times higher than the second group. This is interesting information for when the Committee wants to give proposals further thought.

In the OSIS tool below, the user can select countries and different LtV buckets to make any comparison – the defaults rates are shown on a quarterly basis. The source data is derived from loan level data of securitization transactions. In these transactions, banks have the option to repurchase loans from the underlying pool of assets which could hide observed defaults. Not all banks choose to repurchase and therefore the user can choose to filter out transactions where any repurchase have taken place.

Free Webinar 16th December 2015 – Fair Valuation of Mortgage Loans

By | News | No Comments

European DataWarehouse (ED) would like to inform you of a free, upcoming webinar on “Valuation of Mortgage Loans based on Loan Level Data” held by the data analytics provider Open Source Investor Services (OSIS) on Wednesday 16 December 2015 at 15:00 CET (14:00 GMT, 09:00 EST). This Webex webinar will focus on:

  • Fair values of loans
  • ABS loan level data provided by ED

Future webinars plan to cover IFRS9 modelling and Spread modelling. If you would like to participate, please register with OSIS at: https://meetings.webex.com/collabs/#/meetings/detail?uuid=MB97Z6FLK70AQTKA4ULSQ0T3HH-KFZB&ucs=email&epwd=5925b316050623796b

Dutch mortgage PD model

By | General, News, RMBS Analysis | No Comments

OSIS has developed a new Bayesian PD and LGD model for residential mortgages to estimate, benchmark and backtest PDs and LGDs in a fast, robust and dynamic framework. The model currently covers and is calibrated on 1.8 million Dutch mortgage loans and will shortly be extended to 8 million residential mortgage loans throughout Europe.Average PD per Dutch transaction

Average PD-estimate per Dutch originators

OSIS has used 60% of all the reported Dutch mortgages at the European Datawarehouse (ED) after OSIS applied 1,000 data quality checks on each individual mortgage loan. With the 1.8 million loans rated by the model, analysts can compare the average PD and LGD of 14 different Dutch Originators.

The model will be automatically recalibrated each quarter once new external and/or internal loan level data (“LLD”) becomes available.

PD CycleRecalibration cycle

The initial calibration is based on 3 years of historical loan level data and approximately 14 million observations. The PD Model achieves an accuracy ratio (AR) of 71% in the latest out-of-sample test.CAP curve

Cap curve comparing estimated defaults at 2014Q3 with realised defaults until 2015Q3.