“Chilly showers today, heavier rain and warmer highs tomorrow”
Browsing through TV channels or reading through the weather forecast, we have all at some point come across this phrase. For all the weather departments of this world, time remains an independent factor to draw such a forecast. But if it is raining today it won’t keep pouring forever. Western and eastern disturbances are the drivers and limiters of these forecasts. Modern weather forecasting uses a staggering amount of technology. Weather stations and satellites work around the clock to collect climate data. Powerful supercomputers insert this data into mathematical equations that make up complex computer simulations. It is the modeled output of the randomness of nature to bring out the best possible outcomes. Any market forecast is a similar deal. Our intention is to predict the most likely outcome, based on these inputs.
Today, businesses have to face shrinking profit margins coupled with demand for quick expansion. A prediction based on the union of these two factors using various analytics and forecasting tools help companies hone their competitive advantage. The industry is expected to make decisions in the spur of the moment. This is possible by counting on intelligent forecast models that funnel together incongruent data points into related pieces of spheres (curves). These models should be stout enough to endorse strategic decisions.
We at DRG often, or rather most of the time, have complex sets of data. Our mission is to envisage an intermediate or future value of that data. An answer to this mystery in a ‘Helium Gas’ environment would be to use a ‘straight line of best fit’, draw an equation, and predict. In other words this is an approach of ‘equal triangles’, which is a technique where a simple quotient of 2 similarly shaped but differently sized right triangles, would be used to craft estimates of the missing or the extrapolated data. In common terminology this approach is very similar to what we call a ‘slope of a curve’. However, this approach does not account for the myriad of random factors that can influence a market, including government, economics, and other social factors. Randomness is what we need to structure out to generate our insights. In other words, it is all about converting ‘abstruse-waffled’ predictions into testable intentions and insights. In forecasting, we look at the trends that the data has and use these trends to help forecast future values or values outside the measured data. Raymond Cattell defines fluid intelligence as "…the ability to perceive relationships independent of previous specific practice or instruction concerning those relationships". A forecast model is resultant of high levels of fluid intelligence, which is a combination of abstract thinking and logical reasoning with a lot of crystalized knowledge, which is the pattern of the historical data along with what we know about rules and relationships among the independent variables involved.
Nobody is a Nostradamus but himself. We believe in becoming better forecasters, and cultivating better decision-making processes. This is what the whole intent of this blog series will be. I would discuss numerous ways of ‘modeling curves’ in my upcoming blogs.