Gill Eapen's Blog, page 72
January 10, 2012
Privacy protection
A recent article (1) describes how an individual's privacy can be assured in a database with a restriction on aggregate queries that are allowed. In a database allowing aggregate cross-sectional queries, information at the element level can be revealed by sub segmentation. The suggested solution provides bounds on queries that return subsets with individual information revelation.
Research in this area has been lagging. Most of the effort is currently expended in the collection, aggregation and reporting of information without sufficient concern for the privacy of such information. There are two fundamental avenues of research in this general area. First, one has to create a mathematical foundation for the determination of the minimum amount of data at the individual level (to be stored) that is necessary and sufficient to complete the task at hand and second, one has to determine decision quality or the ability to use all available information in the decision process as a function of stored data. There have been two weaknesses in this process. There is a tendency to collect whatever data are available first and ask questions later as the cost of collection and storage decline fast (2) and most of the stored information is never used in the decision process. This has created problems not only for reaching better decisions but also in the protection afforded. This is because, highly segmented datasets provide less protection against accidental revelation of information as well as planned ones.
Better designs are needed in the definition of necessary and sufficient data, decision processes that operate on such data to improve decision quality and databases that automatically prevent information revelation.
(1) Protecting confidential data with math. Published: Friday, December 16, 2011 - 12:37 in Mathematics & Economics. Source: Society for Industrial and Applied Mathematics
(2) Flexibility: Flexible Companies for the Uncertain World. http://www.routledge.com/books/details/9781439816325/
Privacy protection
A recent article (1) describes how an individual's privacy can be assured in a database with a restriction on aggregate queries that are allowed. In a database allowing aggregate cross-sectional queries, information at the element level can be revealed by sub segmentation. The suggested solution provides bounds on queries that return subsets with individual information revelation.
Research in this area has been lagging. Most of the effort is currently expended in the collection, aggregation and reporting of information without sufficient concern for the privacy of such information. There are two fundamental avenues of research in this general area. First, one has to create a mathematical foundation for the determination of the minimum amount of data at the individual level (to be stored) that is necessary and sufficient to complete the task at hand and second, one has to determine decision quality or the ability to use all available information in the decision process as a function of stored data. There have been two weaknesses in this process. There is a tendency to collect whatever data are available first and ask questions later as the cost of collection and storage decline fast (2) and most of the stored information is never used in the decision process. This has created problems not only for reaching better decisions but also in the protection afforded. This is because, highly segmented datasets provide less protection against accidental revelation of information as well as planned ones.
Better designs are needed in the definition of necessary and sufficient data, decision processes that operate on such data to improve decision quality and databases that automatically prevent information revelation.
(1) Protecting confidential data with math. Published: Friday, December 16, 2011 - 12:37 in Mathematics & Economics. Source: Society for Industrial and Applied Mathematics
(2) Flexibility: Flexible Companies for the Uncertain World. http://www.routledge.com/books/details/9781439816325/
January 6, 2012
The church of pharmaceuticals
A recent study in the Journal of Consumer Research from the University of Chicago Press (1) demonstrates that people hold management policies of churches and pharmaceutical companies to a higher standard. For example, outsourcing of priestly activities by the church to become more efficient or actions to maximize shareholder value by pharmaceutical companies such as hiking the price of well known drugs meet with sharp disapproval of the public, even though such actions are routinely taken by companies in most other industries without any objections or fanfare. So, public, in general, will vote to control the prices of drugs without knowing that such actions will slowly erode the ability of the industry to bring new products to market.
This shows that perceptions and biases are important factors that generate public opinion. Democratic outcomes, thus, are not objective – they are but a reflection of the existing belief systems. Challenging such belief systems, generally, is not a winning formula as humans have shown little ability to change their biases in the presence of data that conflict with belief systems. In many cases, conflicting data reinforces false beliefs. Equally important is the need and desire to conform. If an individual knows that a higher share of the society believes in something, it is more likely that she will conform rather than object.
Education systems around the world has to reform significantly. The best part of education is learning to challenge status-quo and to use logic to study belief systems. Without such skills, the next generation will remain prisoners to arbitrary beliefs perpetuated by the generations before them.
1. From the Commercial to the Communal: Reframing Taboo Trade-offs in Religious and Pharmaceutical Marketing. A. Peter McGraw, Janet A. Schwartz, Philip E. Tetlock
January 3, 2012
Hyperactive Fed
The recent revelation that the Fed is now going to broadcast anticipated future interest rate policies has invigorated the economists in the street and elsewhere as they wasted no time "predicting" the impact of the change. One has to say it is a welcome change from the previous regime when rock star monetary policy makers lived to shock and awe the market at every available opportunity. The biggest of them all, had argued that to get the "right effect," one has to make a policy change that is completely unexpected. And, he certainly got the effects, but not the ones he expected and the regime led a whole generation into despair.
Now, a more intelligent and educated policy maker has initiated a regime of transparency – the exact opposite of what was practiced before. This brilliant insight has been with us for over 50 years now – the idea that reducing uncertainty about future interest rates has the highest value for the economy. As universities around the world shower PhDs on every conceivable empirical work, they miss one important idea – insights are more important than eating data for breakfast, lunch and dinner. The policy makers of yesteryear have indeed been very proud of their empiricism – as they regressed and analyzed every piece of noise that was unearthed from the guts of the economy.
Having a stable interest rate regime and broadcasting it to the market is a good policy. However, there is even a better policy – just let the market know what your money targets are, shut the doors behind you and go home. Spend time with your children as we already know the enigmatic tweaking that goes on behind the walls in Washington and a whole industry that makes their living forecasting and predicting what the Fed will do as they scavenge the streets littered with corpses of small businesses, unable to cope with the uncertainty, destroy significant economic value.
It is time to return to simplicity. Markets are more efficient at setting interest rates. The likelihood of a dozen bureaucrats knowing more is unlikely – if they did , they would not be in their jobs.
January 1, 2012
The certain event
Most of Physics is littered with events that can only be imputed a probability of happening. However, there is an event in contemporary Physics that is almost certain to happen – one of those rare things with a probability of 1.0. It is called the discovery of the Higgs boson particle at CERN – the large hadron collider. It has been shown that scientists have a bias toward proving the hypothesis, once it is stated. As the world waits, a trillion experiments are performed and an inexplicable amount of data is already collected. If the illusive particle is not hiding in the data already extracted, a trillion more collisions can be performed again, ever diminishing the probability of not finding it.
Humans have shown a weakness all through history – they seem to always find what they are looking for. When they stood up in the African savannah, they heard noises from the heaven. The village elder predicted the arrival of further noises, and the followers heard them. Later, religious leaders will predict the arrival of extraterrestrial entities and like clockwork they arrived. Much later, scientists, who shun such nonsense will state hypotheses and then find experimental noise to prove them. Although they hate to admit it, this is also in the grand tradition of the past. They smash things around to show themselves off. They write incomprehensible equations, create inexplicable hypotheses and set out to prove them in grand experiments. And, experiments always confirm the hypotheses.
Some say it is more exciting if they did not find what they set out to find. They will be disappointed – careers are already made and Nobel prizes are awaiting for those who are certain to confirm that the needle does exist in the abundant haystack of noise.
December 30, 2011
Decision ego
A recent study (1) claims that people use logic in making decisions, even though in many cases it may appear that they are going with the gut. The article argues that if the information available is not clear, they may make a gut based call but they are aware of the fact. The study may have to extend further into who makes the decision and how the decisions may impact the decision-maker to fully appreciate the drivers of the decision.
For example, certain decision-makers are paid large sums of money because either the shareholders or the owners of the capital believe that they are good decision-makers. If the results of the decision are not going to be measured for a long period of time or if the results cannot be directly attributed to the decision made, such decision-makers have an incentive to make only gut based decisions. Since their compensation is directly proportional to the untestable belief that they are good decision-makers, it is important for them to shun logic and demonstrate that the decisions they make are private and proprietary.
The decision ego, the need to make gut based decisions, is slowly killing companies and whole industries. Archaic compensation systems and misguided perceptions have created a class of decision-makers, largely driven by ego. In many cases, such repeated experiments make them believe they are in fact good at making gut based decisions and logic does not play much of a role. In most cases, they have to only demonstrate that their decisions are fully private and not that such decisions are good ones. For example, in the venture capital industry, portfolio managers make gut based calls based on what they believe. This is crucial as they have also trained the capital providers to believe that only they can make such decisions.
Decision ego – the propensity to make gut based decisions without analysis or logic – is now prevalent in most industries. Most of this is driven by compensation systems that do not reflect value and perceptions that are perpetuated without logic or analysis (2).
(1) People don't just think with their guts; logic plays a role too. Published: Thursday, December 29, 2011 - 14:32 in Psychology & Sociology
(2) Decision Options: The Art and Science of Making Decisions. http://www.crcpress.com/product/isbn/9781420086829
December 29, 2011
Impersonal tests
Personality tests – nearly a billion $ industry – have been used routinely by employers and universities. They are supposed to reveal the true orientation of an individual – extrovert v/s introvert, perceiving v/s judging and many other such concocted dimensions of personality. Humans always liked characterizing things neatly into buckets. Initially, they used observable characteristics such as color of skin and eyes, weight and height and body odor. As they became more "sophisticated," more intangible but measurable attributes were used for the same – including ethnicity, nationality, religion, political affiliation and blood type. Later, this urge to characterize, took them to invent horoscopes and personality tests.
The personality test, thus, is yet another segmentation scheme invented by humans, with a robust and rich history of segmentation ever since they arrived on earth. Every human from inception had to endure rigid segmentations in many dimensions – each put down with elaborate definition and color. They divided people into the clergy and the warriors, financiers and farmers, the rich and the poor and the religious and the not so religious. They labeled people according to where they chose to live – some in Wall street, others in Main street and yet others simply street less, some in the cities, others in the suburbs and yet others in the farms, some in the East, others in the West and those in the middle. Segmentation has been the only enduring core competence of humans and they are getting ever better at it with every passing day.
The personality test, a more advanced segmentation scheme, are well liked by those who use it. Companies like "leaders," and some personality tests show clearly whether the next impending brick in the wall can be marked as such or sent back to the crusher. Universities like "thinkers," as they set out to groom the next crop of Nobel laureates from the raw and uncut inputs they are given and the personality test is an invaluable tool to select the most likely. Nuclear power plants and aircraft control towers seek those who can remain calm under stress and those who are balanced. Venture capitalists seek the creative types, who can not only create but also ride the bus to the next exit before the ink in the investment check dries. Investment bankers seek the "money types," who can simply make money without any consideration for humanity and personality tests reveal this as well. Personality tests have become an important tool in many areas of selection, design and management. However, there is little evidence that the labels invented by these arbitrary tests have any correlation to what they are used for. What may be missing in them is the person, herself, who is rendered incompetent by the tests administered on her.
Humanity should celebrate the complexity of the individual and not their ability to stuff everybody into a 4 x 4 matrix.
December 20, 2011
Descale, now!
Management gurus and consulting moguls have always been high on scale. Scale, they argue, reduces costs and increases efficiency. Venture capitalists, thought to be the third rail of the economy, mumble the two magic words - "scale" and "exit" in deep sleep, for without it, they cannot be interested in investing. Business schools teach every budding manager and entrepreneur, the magic of scale – how profits can increase exponentially and how wealth can be maximized under the spell of scale.
Scale is a misguided notion from the industrial revolution, when it was indeed valid. Modern companies are unable to break the shackles of the past – as their managers are educated in the same ideas and their financiers are still left with limited brain cells. The analysts of the street, with wisdom that cannot be measured with any modern techniques, analyze and reanalyze how scale is helping their fledging companies as they stuff supply chains to dress up next quarter's earnings. Accountants jump up in sheer joy as they witness numbers add up horizontally and vertically in spreadsheets and they are driven to orgasm when they actually tie everything together for everybody to see and enjoy. Scale, still, is the driving notion of business success.
It is time companies seriously considered descaling as a strategic need (1). Contrary to popular opinion, size does not really matter in the information economy.
1. Flexibility : Flexible Companies for the Uncertain World
http://www.crcpress.com/product/isbn/9781439816325
December 19, 2011
True permanence
A recent study (1) casts doubts on the idea that has been universally accepted – babies, less than a year old, have difficulty with object permanence. When an object is shown to a baby and then hidden, babies below certain age, have shown an inability to assume that the object actually exists. The study shows that such a conclusion is premature and not necessarily true. Using an experiment in which the object's shape was changed when it appeared again from behind a screen and measuring the time the babies spent glancing at it, the study demonstrates that babies actually have pointers in their brain for object permanence.
This is interesting. It can lead to a broader conclusion that most of the conceptual competence of the human brain is in place at birth. The operating system, itself is capable of most of the complex tasks the brain will perform as the baby ages. The results are different later only in that the brain has acquired knowledge and there may not be any improvement in its innate capabilities. Conceptually, such a finding, may also have implications for answering the question of when life begins. If the brain is delivered to the world, with a fully functioning operating system, then it means that such capabilities are inherent at a time before birth. The question will be how far before birth the baby possesses a conceptually capable brain.
Policy in this area has been controversial. Those who dismiss such policy questions either because of ignorance or for convenience may be taking it too lightly.
(1) Babies remember even as they seem to forget, Published: Monday, December 19, 2011 - 15:26 in Psychology & Sociology
December 16, 2011
The need for ignorance
A recent study shows that some percentage of uninformed (ignorant) individuals promote democratic consensus in animal groups. This is a very interesting finding as the same appears to be true in complex human societies as well. Through theory and experiments, the article demonstrates that the presence of uninformed individuals inhibit the process of domination by a strongly opinionated minority. If the strongly opinionated minority is pushing toward an optimal outcome for the system, ignorant participants will slow them down.
These findings have a variety of societal and policy implications. At the limit, in a system with fully informed individuals, strongly opinionated minority is more able to push the outcomes in their desired direction better. On the other side, a system with complete ignorance will assure democratized outcomes, regardless of the optimality of such outcomes. Most systems, fall in between and it is unclear what may be the best level of ignorance that should be maintained to push the system to the best outcome.
A proxy for informed systems may be large and complex companies, where an opinionated minority drives the agenda. An example of uninformed systems may be large democracies that generally assure democratic outcomes. Since neither democratic outcomes nor agendas driven by the opinionated minority can be shown to be optimal, it is difficult to determine the best level of ignorance. However, it is likely that both a complete ignorance or a lack of it, result in inefficient outcomes.
1. Uninformed Individuals Promote Democratic Consensus in Animal Groups
Science 16 December 2011: Vol. 334 no. 6062 pp. 1578-1580 DOI: 10.1126/science.1210280
Iain D. Couzin1,*, Christos C. Ioannou1,†, Güven Demirel2, Thilo Gross2,‡, Colin J. Torney1, Andrew Hartnett1, Larissa Conradt3,§, Simon A. Levin1,
Naomi E. Leonard4
