top of page
  • Writer's pictureAdmin

B1g d4t4 and I0T

Old paradigms of language do not apply to this field ;-). I recently had the privilege to attend an industrial engineering conference that focused on utilising large quantities of data, connected devices and sensors to improve manufacturing and support. I am familiar with a few buzz words juggled about in this field. Post the conference I definitely know that there is much to learn of this nascent ecosystem.

Speakers at the conference highlighted how they used "big data" and "IoT" (internet of things) in making their operations more efficient, notable speakers were from Hitachi and Nasdaq. Some companies were storing and analysing in excess of 15 terabytes of data per day, now that's big. The problem with buzz words are they compress years of effort into a word that then goes viral and loses its essence. What I got from the actual practitioners is that big data, machine learning, AI and IoT are many times more complex to integrate and implement than what the catch phrases portend, much of it I don't know but I am making some efforts in the field. Use cases and field experiments presented were astounding, from agile manufacturing to predictive maintenance. A few things stood out for me, the parts I understood, were data governance and customer 360.

Before any advanced analytics can be performed a data governance framework must be adopted. Briefly the framework must include procedures that govern collection; transformation; storage; retrieval; presentation and analytics of data. The concept of a data lake was introduced. Speakers advocated including internal (company generated data) with data from external sources in such a lake. Hoarding and building a very large database beyond requirements was the message from the practitioners.

I was familiar with the concept of a customer 360 but overlaying very large data sets and advanced analytics just augments the customer 360 greatly. A customer 360-degree is the idea that companies can get a complete view of customers by aggregating data from the various touch points that a customer may use to contact a company to purchase products/services and receive support. The customer can be internal or external. I believe integrating service design (an 80's concept) with a customer 360 will put a brand front and center of consumers' hearts and minds. The difficulty is building the analytical model and knitting the data. I have seen disastrous uses of data manipulation to gain approval for vanity projects.

Other advanced components were discussed like hadoop, propogation, spark interface, etc. Unfortunately I don't know much about those. However the overall theme was clear to me, the advances in hardware and software are being used to ameliorate the physical and mental shortcomings of people. Pundits call it the 4th industrial revolution, well progress is marked by invention. Many theorists and consultants either have a dystopian or utopian future scenario but in reality I am sure we will end up somewhere in the middle. Future economies need to plan for a world of ever increasing supply with the possibility of prolonged disinflation, demand is going to be problem. The idea of trickle down economics needs to change and so should the distribution mechanism of growth, which is exceptionally complex to theorise without reducing it to capitalism or communism. Also with technology progress outrunning social progress the idea of central bank controlled money and government debt needs deep thinking and scenario analysis. The finance sector is seeing technology challenge the status quo.

In the banking and investing field advances in technology has broadly segregated the industry into 2 classes, I refer to them as "hunters" and "gatherers". Old banks are seeing competition from alternative payment systems that are much faster and cheaper. By using the latest in secure payment tech protocol these platforms have lower costs than the legacy network of people and processes that banks possess. The platforms serve to gather more and more transactions with a sliver of margin over the cost of cloud server time. Hunters use deep learning tech to quantify credit, market and liquidity risks thereby creating advanced peer to peer lending networks (for example crowd funding). These platforms are compensated for the high effort through higher margins. This threatens the banks' model of borrowing very low and lending very high. Central banks refer to alternative (non-bank) payments and lending systems as "shadow banking".

Advances in software and protocols have also allowed for straight through processing between exchanges, custody, clearing agents, back offices etc. which minimises a fund's operational considerations. Gatherers (ETFs and passive mutual funds) use the tech improvements to accumulate assets under management by passively managing assets through index tracking thus maintaining profit by keeping costs and margins ultra low. Hunters (hedge funds and private equity) use the tech for high speed trading, deep pattern analysis and information acquisition. Here it is a high cost and high effort game therefore the margins are much higher. Traditional asset managers, which don't have much of an identity now, have seen their market share erode.

One can easily search online for big data, AI and IoT research and applications but also try to attend an industry/practitioners workshop or 2. Try to avoid the big box accounting consultants seminars they just reduce to catch words. This field is complex thus its best to get first hand experience from people who actually implemented.

Happy trading!

*I wrote this article myself and it does not constitute advice or trade recommendations.


bottom of page