Tuesday, July 4, 2017

Saturday, April 8, 2017

Back to the public-vs-private debt debate

Back to the public-vs-private debt debate for beleaguered SA parastatals: 

"This scenario would mirror the events of 1985 when then President PW Botha delivered his crazed Rubicon Speech which caused a $13 billion default. The foreign debt/GDP ratio hit 40%. Today it is nearly 50%. Botha was compelled to impose exchange controls."


Then there are perspectives from bigger markets:


Sunday, April 2, 2017

Which ZARandlord are you ?




Reduced JSE fees have a cost
Regulatory changes aimed at encouraging very fast, technology-driven trading on the JSE may have some unintended consequences.
Global markets have undergone several transformations since the repeal of the Glass-Steagall act in 1999. This enabled US-based global investment banks to trade on behalf their own proprietary interests, as well as the interests of their customers.
While the JSE is technically an older, medium sized market by capitalisation in the global network, its electronic trading platform in Johannesburg and fee model are relatively new.
In September 2013 the JSE overhauled its transaction billing system to eliminate the minimum fee of 4.00 ZAR per trade (excl. VAT), allowing fees to comprise a significantly smaller part of single-transaction commissions. At least one anticipated benefit was to make algorithmic buying and selling of large orders feasible by facilitating strategies of smaller, less obvious trades.
In addition to this, the exchanges matching system [computer hardware and software], was brought back from London to South Africa and a new co-location service on the JSE’s premises, for investors to house their computerised trading systems, were announced.
“These regulatory changes were aimed at encouraging very fast, technology-driven trading on the JSE,” says Prof Diane Wilcox of the QuERI Lab Research group in the School of Computer Science and Applied Mathematics at the University of the Witwatersrand. In developed markets, minimal transaction costs per trade make the processing of trillions of orders per day possible. Like any large-scale, safety-critical system, an electronic trading platform is vulnerable to process error and 'viruses'. For markets, the latter can include trading strategies which abuse technological advantage to extract risk-free profit or manipulate prices. If a market enables too many predatory trading algorithms then capital is leaked away from optimal investment and information contained in prices becomes less reliable."
Wilcox and her colleagues, Prof Tim Gebbie and graduate students Mr Michael Harvey and Dr Dieter Hendricks, presented their research on unintended consequences of the regulatory changes at the JSE in a recent paper on deviations in price impact.
For sustainability, market makers need to ensure market quality to attract and protect investment, and hence, the price-formation process. To this end, various measurements can be monitored. In 2003, a team of physicists reported on an investigation in the journal Nature, which applied advanced dimensional analysis to large volumes of data from the New York Stock Exchange. They uncovered evidence for a new demand- supply relationship which related price-change to volume of transaction.
Persistent deviations in price impact and other market quality estimators can be symptomatic of the presence of disruptive arbitrage,” say Wilcox. Thus, tracking such measurements is similar to monitoring the health of a patient or high-performance athlete. Significant changes in expected levels can be used to halt trading under extreme conditions.
Major market participants, who now rely on mathematical algorithms rather than human traders, have acquired more computing power and software development to stay competitive, but these costs are still passed on to the consumer ,” says Gebbie.
The work of the QuERILab team reports on the occurrence of an anomaly in the price impacts of small transaction volumes for financial stocks. It was found that the price impact corresponding to smaller volume trades was greater than expected, relative to prior estimates for stocks comprising the FINI index. Such deviations can have significant implications for risk management. The research shows how a decrease in direct transaction costs resulted in an unexpected increase in indirect costs through price impact.
“Our investigation is aimed at improving the understanding market microstructure. This refers to the combination of regulation and expected properties of data recorded in a high-precision electronic order book of different types of orders and transactions which match buyers and sellers ,” says Wilcox.
Empirical studies of price formation suggest that regulatory changes, particularly those encouraging varieties of technology-driven races associated with speed of trading and increased volume, may have unintended consequences for costs to end-users, such as pension funds, by introducing novel risks within markets or increasing market fragility. 
Some stakeholders support the view that competition between more local stock exchanges offers insurance against these new risks, while others advocate for more regulation in public interest.
An inevitable part of the “arms race” at stock markets around the world is an increased need for skilled risk measurement and management in financial markets. Where previously traders coped with astute, quick calculations on noisy trading floors, the transformation to electronic order books brought a shift in the required quantitative skills for understanding price formation.
“The application of high performance computing to analyze large volumes of historical data and respond to new information to make trade decisions within milliseconds [0.001 of a second] or microseconds [0.000001 of a second], has caused a massive disruption in the industry,” says Gebbie.
“It is a bit like what Uber has done to parts of global taxi industry, but in the trading world. Experienced traders, who needed business savvy and some asymmetric access to market information – perhaps based on social networks and skill accumulated over many years, have been and are being replaced by scientists and engineers, who combine analytic skill and knowledge of the technological developments in markets to optimise allocation of capital,” says Gebbie.
About the research
The research was based on the analysis of terabytes of transaction data, sourced from Thomson-Reuters, and has been funded by the National Research Foundation as part of a broader research project, aimed at better understanding financial markets.
The QuERI Lab research group supports research-driven teaching in the Computational Finance BSc Honours courses within the School of Computer Science and Applied Mathematics at WITS University. Find the research paper online here.
- See more at: http://www.wits.ac.za/news/latest-news/research-news/2017/2017-03/reduced-jse-fees-have-a-cost.html#sthash.ySSCCpU0.dpuf

Thursday, February 9, 2017

(Interesting Books) Why Nations Fail (2013) Acemogly & Robinson

A thought provoking read highlighting why "trade-test innovation" is at the heart of economic
success when states are "inclusive" rather than "extractive". That an organized and centralize state may be a critical component for economic success but that the centralization must avoid the path to extractive economics by an elite, and balance the need for centralization against interests of societies members. That the incentives created by inclusive organization seem to generate the associated trade-test innovation necessary for the creative destruction that has historically delivered economic growth. Creative destruction that perpetually threatens the controls by the elites of any organized and centralized state, but through trade and innovation creates the opportunities for a better and more prosperous life for those in broader society.

Why Nations Fail (2013) ACEMOGLU AND ROBINSON


" ... Why Nations Fail answers the question that has stumped the experts for centuries: Why are some nations rich and others poor, divided by wealth and poverty, health and sickness, food and famine?

Is it culture, the weather, geography? Perhaps ignorance of what the right policies are?

Simply, no. None of these factors is either definitive or destiny. Otherwise, how to explain why Botswana has become one of the fastest-growing countries in the world, while other African nations, such as Zimbabwe, the Congo, and Sierra Leone, are mired in poverty and violence? ..."

SAFA 2017 Talk : 18-20 January 2017 Cape Town, South Africa

Title: Learning zero-cost portfolio selection with pattern matching for intraday trading

Speaker: Tim Gebbie

Authors: Tim Gebbie, Fayyaaz Loonat

Abstract: We consider and extend an adversarial agent-based learning approach to the situation of zero-cost portfolio selection in the domain of quantitative trading. The algorithms are directly compared to standard NYSE test cases from prior literature. The learning algorithm is used to select parameters for agents (or experts) generated by pattern matching past dynamics using a simple nearest-neighbour search algorithm. We demonstrate that patterns in financial time-series on the JSE can be systematically exploited in collective, but that this does not imply predictability of the individual asset time-series themselves. We show that these types of machine learning algorithms are well suited for intra-day quantitative trading.



Monday, November 28, 2016

Can Agent-Based Models Probe Market Microstructure? Donovan Platt, Tim Gebbie (2016) arXiv.org


Can Agent-Based Models Probe Market Microstructure?
Authors: Donovan Platt, Tim Gebbie
Abstract: We extend prior evidence that naively using intraday agent-based models that involve realistic order-matching processes for modeling continuous-time double auction markets seems to fail to be able to provide a robust link between data and many model parameters, even when these models are able to reproduce a number of well-known stylized facts of return time series. We demonstrate that while the parameters of intraday agent-based models rooted in market microstructure can be meaningfully calibrated, those exclusively related to agent behaviors and incentives remain problematic. This could simply be a failure of the calibration techniques used but we argue that the observed parameter degeneracies are most likely a consequence of the realistic matching processes employed in these models. This suggests that alternative approaches to linking data, phenomenology and market structure may be necessary and that the stylized fact-centric validation of intraday agent-based models is insufficient, and warns that increased mechanistic complexity of agent-based market models may lead to flawed insights.

Comments: 15 pages, 8 figures
Subjects: Computational Finance (q-fin.CP); Trading and Market Microstructure (q-fin.TR)
Cite as: arXiv:1611.08510 [q-fin.CP]
 (or arXiv:1611.08510v1 [q-fin.CP] for this version)

Sunday, November 27, 2016

On regime change and hierarchical causation

60 years ago in Nov 1956, a small island economy took the first giant steps towards balanced education, health and significant reduction of local inequality, adaptively addressing economic sustainability thereafter. The Granma news cover from 1986 challenges perspectives with respect to the quantification of emergence. RIP Prof Castro.