for Decision Making
at the Frontiers of Innovation
A cascade of advances in data science coupled with hardware technology have given rise to newborn techniques to uplift the efficiency and effectiveness of decision making in knotty domains. The areas of application span the spectrum from scientific research and engineering design to medical diagnosis and investment strategy.
On the downside, however, the melange of tools in use today falls far short of the ideal. The deficit is especially dire in messy domains bubbling with garbled data and fluky events. A showcase lies in the financial forum, a field so roily and slippy that it confounds the mass of participants ranging from part-time amateurs to full-time professionals.
As an example, the drove of putative gurus in the stock market are unable to perform even the simplest task of prediction: namely, foretelling whether the bourse will rise or fall going forward. In other words, the pundits as a group fail to match the flip of a coin in flagging the direction of the market.
In terms of active strategies for investment, the very attempt to outwit the market leads the bulk of investors astray. By second-guessing their own judgments along with the findings of their analytic tools, the actors in toto – be they greenhorns or elders – end up lagging behind the passive benchmarks of the market. In a stark irony of human affairs, the financial tract is so miry and shifty that the mounds of time and energy expended by millions of hustlers round the world are not only feckless and worthless but downright counterproductive and harmful to their own cause.
In seeking a better alternative to the morass, a faithful course lies in a wholesome strategy that combines the strengths of motley techniques while bypassing their attendant pitfalls. For this purpose, the building blocks run the gamut from robust statistics and factor analysis to neural networks and genetic algorithms.
A second and related earmark of a cogent approach involves the proper use of extant tools from the armory of data science. As a backdrop, the muddle of myths and mistakes in the marketplace springs from patchy data and sketchy samples mashed up with faulty logic. In that case, a forthright cure lies in the methodic uptake of the tools from data science. An example involves the incisive use of statistical tests to verify the tentative conclusions reached by intuitive means or rational schemes.
No wonder that the crush of supposed experts in the field have a long-standing custom of shooting themselves in the foot. A fine example involves the pundits in the aggregate – whether the practitioners in the trenches or the researchers on the sidelines – who rely on scrappy data and succumb to woozy logic while brushing aside the facts of life in the real economy as well as the nubs of common sense in the financial realm. Amid the thrashing and fumbling, the soothsayers rightly question their own methods and wrongly issue forecasts with awesome consistency.
As noted earlier, a fitting tack is to employ a suite of statistical tests to confirm the apparent patterns that emerge in baffling fields. A checkup of this kind may serve as a bulwark against the deluge of myths and muffs. A second and related boon is to quell the compulsion for second-guessing that foils the mass of decision makers in cryptic domains.
All too often, the digital techniques and software systems in use today fail to live up to their promise. On a cheery note, though, a trenchant tool crafted with care may outperform its own creator as well as the seasoned expert in the field. An example lies in medical diagnosis where a smart robot can peruse an X-ray image and conduct an assay with greater proficiency than a human radiologist.
For the sake of concreteness, the financial forum represents a multifaceted showcase for exploring the challenges and opportunities in store. The realm of investment serves as a proving ground of utmost complexity and difficulty for all manner of decision makers. The action in the marketplace springs from a slew of drivers ranging from economic conditions, product innovations, and psychic forces to fiscal policies, geopolitical forces, and natural disasters. Despite the focus of the case studies at the outset, though, the basic concepts and techniques under study are applicable to the panoply of opaque domains.
To sum up, a hybrid model for decision making, properly designed and employed, can outshine any of its constituent parts. For this reason, the mosaic approach may uphold a new generation of supple tools to distill useful knowledge from noisy data and construct versatile bots for tricky tasks. The ductile systems of this breed may support human principals on the fly as well as automate toilsome jobs in background mode. The upshot is bound to be a flowering of efficiency and effectiveness, of ingenuity and productivity, in abstruse fields of all kinds.