arXiv subjects

Andrew Green

Publications and source records attributed to Andrew Green.

24 records · Page 2Linked to original sources

arXiv2013

Collateral-Enhanced Default Risk

Changes in collateralization have been implicated in significant default (or near-default) events during the financial crisis, most notably with AIG. We have developed a framework for quantifying this effect based on moving between Merton-type and Black-Cox-type structural default models. Our framework leads to a single equation that emcompasses the range of possibilities, including collateralization remargining frequency (i.e. discrete observations). We show that increases in collateralization, by exposing entities to daily mark-to-market volatility, enhance default probability. This quantifies the well-known problem with collateral triggers. Furthermore our model can be used to quantify the degree to which central counterparties, whilst removing credit risk transmission, systematically increase default risk.

q-fin.RM↗

arXiv2012

Will Central Counterparties become the New Rating Agencies?

Central Counterparties (CCPs) are widely promoted as a requirement for safe banking with little dissent except on technical grounds (such as proliferation of CCPs). Whilst CCPs can have major operational positives, we argue that CCPs have many of the business characteristics of Rating Agencies, and face similar business pressures. Thus we see a risk that prices from CCPs may develop the characteristics attributed to ratings from Rating Agency pre-crisis. Business over-reliance on ratings of questionable accuracy is seen as a cause of the financial crisis. We see the potential for same situation to be repeated with prices from CCPs. Thus the regulatory emphasis on CCPs, rather than on collateralization, may create the preconditions for an avoidable repeat of the financial crisis.

q-fin.GN↗

arXiv2012

CDS pricing under Basel III: capital relief and default protection

Basel III introduces new capital charges for CVA. These charges, and the Basel 2.5 default capital charge can be mitigated by CDS. Therefore, to price in the capital relief that CDS contracts provide, we introduce a CDS pricing model with three legs: premium; default protection; and capital relief. If markets are complete, with no CDS bond basis, then CDSs can be replicated by taking short positions in risky floating bonds issued by the reference entity and a riskless bank account. If these conditions do not hold, then it is theoretically possible that the capital relief that CDSs provide may be priced in. Thus our model provides bounds on the CDS-implied hazard rates when markets are incomplete. Under simple assumptions we show that 20% to over 50% of observed CDS spread could be due to priced in capital relief. Given that this is different for IMM and non-IMM banks will we see differential pricing?

q-fin.PR↗

arXiv2012

AAO Observer Number 122 (August 2012)

The newsletter of the Australian Astronomical Observatory. In this issue: SPIE Extravaganza; The AAO's Gemini High-resolution Optical SpecTrograph (GHOST) concept; First Cosmological Constraints from the 6dFGS Peculiar Velocity Field; Galaxy And Mass Assembly (GAMA): GAMA Announces Second Data Release; Evidence for Significant Growth in the Stellar Mass of the Most Massive Galaxies in the Universe; The 5th Southern Cross Conference: A Joint CASS/AAO Conference; GALAH: Preparing for Flight; and all the usual columns and news from the Observatory.

astro-ph.IM↗

arXiv2012

The Tully-Fisher Relation for 25,000 SDSS Galaxies as Function of Environment

We construct Tully-Fisher relationships (TFRs) in the $u$, $g$, $r$, $i$ and $z$ bands and stellar mass TFRs (smTFRs) for a sample of $25,698$ late spiral type galaxies (with $0.045 0.045$) the fibres sample enough of the disc to obtain a linear relationship between $v_{\rm FWHM}$ and $v_{\rm circ}$, allowing us to obtain a TFR and to investigate dependence on other variables. We also develop a methodology for distinguishing between astrophysical and sample bias in the fibre TFR trends. We observe the well-known steepening of the TFR in redder bands in our sample. We divide the sample of galaxies into four equal groups using projected neighbour density ($Σ$) quartiles and find no significant dependence on environment, extending previous work to a wider range of environments and a much larger sample. Having demonstrated that we can construct SDSS-based TFRs is very useful for future applications because of the large sample size available.

astro-ph.CO↗

arXiv2001

MLAPM - a C code for cosmological simulations

We present a computer code written in C that is designed to simulate structure formation from collisionless matter. The code is purely grid-based and uses a recursively refined Cartesian grid to solve Poisson's equation for the potential, rather than obtaining the potential from a Green's function. Refinements can have arbitrary shapes and in practice closely follow the complex morphology of the density field that evolves. The timestep shortens by a factor two with each successive refinement. It is argued that an appropriate choice of softening length is of great importance and that the softening should be at all points an appropriate multiple of the local inter-particle separation. Unlike tree and P3M codes, multigrid codes automatically satisfy this requirement. We show that at early times and low densities in cosmological simulations, the softening needs to be significantly smaller relative to the inter-particle separation than in virialized regions. Tests of the ability of the code's Poisson solver to recover the gravitational fields of both virialized halos and Zel'dovich waves are presented, as are tests of the code's ability to reproduce analytic solutions for plane-wave evolution. The times required to conduct a LCDM cosmological simulation for various configurations are compared with the times required to complete the same simulation with the ART, AP3M and GADGET codes. The power spectra, halo mass functions and halo-halo correlation functions of simulations conducted with different codes are compared.

astro-ph↗