Statistical analysis
Podcast: Piterbarg and Nowaczyk on running better backtests
Quants discuss new way to extract independent samples from correlated datasets
Quantum kit offers HFTs ‘100-fold’ speed boost
After spotting FX arbitrage opportunities, new tech faces real-world test in Japanese stocks
A hybrid model for credit risk assessment: empirical validation by real-world credit data
This paper examines which hybridization strategy is more suitable for credit risk assessment in the dynamic financial world.
An alternative statistical framework for credit default prediction
This study compares the gradient-boosting model with four other well-known classifiers, namely, a classification and regression tree (CART), logistic regression (LR), multivariate adaptive regression splines (MARS) and a random forest (RF).
Harnessing AI to achieve Libor transition
Chris Dias, principal at KPMG, explains how the vast increase in accuracy that artificial intelligence (AI) offers when dealing with large volumes of complex agreements is crucial to exploring the market opportunities and mitigating the risks of the…
A statistical technique to enhance application scorecard monitoring
Application scoring plays a critical role in determining the future quality of a lender’s book. It is therefore important to monitor the performance of an application scorecard to ensure it performs as expected.
The implicit constraints of Fundamental Review of the Trading Book profit-and-loss-attribution testing and a possible alternative framework
This paper presents the constraints embedded in the the profit-and-loss-attribution test and explores a possible alternative framework.
Cyber modelling masks scale of potential losses, study finds
Different statistical approaches produce big variations in future loss estimates, says Esma researcher
New statistical approach proposed to tackle internal fraud
Tests improve on methods to identify anomalous data created by fraudsters
Quants head for the shop floor
Demand for technical skills is growing, but roles have changed – and some schools are not keeping up
Cutting Edge introduction: Creative stress testing
New stress-testing method offers a break from decades-old traditio
Winton sees problems with big data analytics
Social media and live newsfeeds not so far useful, hedge fund says
The simple link from default to LGD
The simple link from default to LGD
Resist the rise of the risk management machines
Overreliance on modern risk management systems, and metrics such as value-at-risk, can blind firms to tectonic structural market shifts. To help alleviate this problem, the use of human judgement and intervention is required, argues Vincent Kaminski
Asia Commodity Research House of the Year: SG CIB
Accurate forecasting and insightful analysis prove a winning formula for SG CIB
Energy companies face software integration issues
Energy Risk software survey shows traders plan to add to ETRM systems in 2011; vendors discuss integration problems
ECB calls for more data to prevent crises
European Central Bank draws attention to information gaps, says broader and more flexible data will aid macroprudential supervision
Modest means
Credit loss models typically calibrate default separate from loss given default. Here, Jon Frye calibrates simultaneously, using credit loss data. This produces a surprising test result: the credit loss models do not significantly outperform a…
Sponsor's article > Statistical process control
Too often, finance professionals manifest a smug sense of superiority towards their peers in manufacturing. In this third column in a series, David Rowe argues that when it comes to operational risk management, the manufacturing sector has much to teach…
Accord preparations: the rest is yet to come
While the debates have raged for months about many aspects of the proposed Basel II Accord, on some points there has been relative silence, in particular with regard to the seeming overreliance on statistical techniques.
Pitfalls and alternatives
Correlation