Netherlands Job Openings

MN

AXP Assignment - Improving back testing environments of financial markets

The Hague

November 1, 2024

AXP
This assignment is part of the unique Academic Excellence Program (AXP). This program is a cooperation between MN and PGGM, two leading Dutch pension fund service providers. Students will get the opportunity to work alongside investment professionals, collaborate with other students and learn about institutional investing. On regular AXP-days all students in the program will come together in The Hague or Zeist to share ideas, learn from each other, attend presentations from investment teams and discuss job opportunities.

Research topic

MN has developed the FX Smart Order Execution Solution, an automated execution engine that can perform algorithmic execution of orders in the FX spot market. Putting a large order in the market all at once could result in unwanted high market impact and high costs, so instead an order – henceforth called a parent order – is divided into multiple smaller child orders. These child orders are then executed according to the specific algorithm chosen by the trader. The algorithms can take orders aggressively out of the order book at a certain price with a market order (resulting in spread costs) or can place limit orders in the order book, passively waiting for the order to be executed at a favorable price (resulting in opportunity costs). MN also developed a backtesting environment to assess the viability of changes in the parameters of the existing execution strategies - e.g. for optimalisation of the level of aggressiveness of the child orders - or a completely new strategy to execute the parent orders. A big challenge of the backtesting environment is to simulate the market reaction of placing, cancelling, updating and filling of the child orders, as this what-if scenarios are unknown in the historical dataset. This especially applies for filling the passive child orders, as it takes an aggressive market participant to trade though the spread and take these passive orders out of the order book. The challenge for this research is to improve the reliability and quality of MN’s backtesting environment by acquiring more comprehensive data, refining simulation assumptions, and developing sophisticated models to assess the market reaction and market impact. An angle that can be considered to approach this backtesting challenge is to explore more sophisticated stochastic models that account for time-varying arrival rates and other factors influencing market order dynamics, such as Hawkes processes. Another approach is to investigate the benefits of applying machine learning frameworks. For example, a generative adversarial network (GAN) could be used to simulate market events in an orderbook by putting two neural networks in context, where a generative network generates fake orders while the discriminative network evaluates them based on a real orders and orderbook data. Both networks learn from the outcomes and keep on improving, leading to the generative network creating fake orders that are very hard to distinguish from real orders and therefore perfect for backtesting purposes. If a stochastic, a machine learning, and/or potentially another approach can be successfully applied on FX orderbook data to improve the backtesting environment, it Classificatie: Vertrouwelijk is also interesting to see if the framework(s) and the insight can be applied in a more general backtesting framework that can be used by MN for multiple assets and instruments.
Data
For this project high density orderbook and tradebook data for FX will be provided.

Your profile

For these assignments we are looking for ambitious students with:

Quantitative background (eg. Econometrics, Mathematics, Financial Engineering, etc.).
Programming skills in Python.

Your work environment

The graduation student will be placed in the Investment Management team within the Treasury department of MN Asset Management. The Treasury department is responsible a.o. for managing the liquidity portfolios and currency overlays of our Dutch pension fund clients. The team comprises of five investment managers, four collateral and securities lending managers, two algorithmic execution specialists, one analyst and the head of Treasury.

MN as an employer

MN is serving 7 pension fund clients in the areas of Board Advisory, Asset Management, and Fiduciary Management. For our clients we manage pension assets worth approximately EUR 150 billion, that makes MN the third largest pension fund Asset Manager in the Netherlands. About 35.000 companies (employers) in various industries are affiliated with our clients, representing about 2 million pension fund participants in total. We invest globally in equities, government- and corporate bonds, loans, real estate, and private equity, etc.

As an MN employee, you can expect

A monthly internship allowance;
Emphasis on work-life balance and flexibility in working hours;
Numerous opportunities for learning and personal development and a calendar filled with social events;
Hybrid working environment as the default, with a modern office building in the center of The Hague. It is easily accessible by public transport and has in-house facilities for exercise, relaxation or a refreshing coffee from our in-house Barista.
The internship should be scheduled within the first half year of 2025. For this assignment the main location will be MN in The Hague.

Contact

We would like to meet you! Click on the apply button and upload your resume and motivation. If you have any questions, please contact Siona van Dijk via recruitment@mn.nl.
We look forward to receiving your application and welcoming you to MN!

New Job Alerts

Big Ass Battery

Heineken International B.V.

Manager Customer Experience - Short Term Assignment

Leiden

FULL TIME

September 3, 2024

View Job Description

Senior Manager MKB Accountancy

The Hague

FULL TIME

November 15, 2024

View Job Description

Graduation: Categorization and Classification of Software Applications in an Automated Warehouse Platform

Veghel

FULL TIME

November 15, 2024

View Job Description