It was a dark and stormy night….no, no – that’s the start of another story. This story starts in Lewiston, Maine, where I was attending Bates College, with plans to obtain a Bachelor’s in mathematics. Prior to my undergraduate work, I had always loved applied math – so I thought that getting a degree in math was the right direction to take. But the math program at Bates was all theoretical – the only numbers used in the classes were the numbers that labeled the steps of the proofs. One semester we actually ran out of Greek letters when solving proofs, so we had to start using Chinese letters to fill the gap.
So I went in search of applied math at Bates – which I found in both physics (where we turned infinity into just a really big number) and psychology. The appeal of psychology was found in experimental design – and there I became fascinated by statistics. The experiments we created were fun to design and test (Do be careful, however, when testing someone’s personal space – one person so disliked the invasion of his space that he hit me with his notebook). I still enjoyed physics (although I have to admit that my lab work in electronics was not overly successful – my soldering skill left something to be desired), so I graduated with an interdisciplinary Bachelor’s degree in psychology and physics.
Having enjoyed my statistical work in psychology, I thought I should pursue a graduate degree in experimental psychology. Off I went to Vanderbilt University in Nashville, to start a Ph.D. program in psychology. Again I had an epiphany – in order to develop data to test statistically, I needed to create and conduct psychological tests. One experiment I created required the ever available sophomore volunteers to deduce the direction of movement of two points of light in space. The experiment was so long (two hours) and difficult that some students cried in frustration. I was making Vanderbilt football players cry! It was then that I decided I wanted to pursue a degree that didn’t result in anyone crying. Using my physics degree as a springboard (people don’t tend to cry in physics), I switched programs at Vanderbilt and started a Master’s in engineering management.
And there in Vanderbilt’s Engineering School I took my first class in forecasting – Technological Forecasting and Assessment. Statistics and applied math together with a dose of scientific fortune-telling! Trend extrapolation, growth curves, the Delphi procedure, cross-impact analysis – with time series data to study and dissect. My love of applied math and statistics gave me a glimpse into the future; I was hooked. The topic of my Master’s thesis was a handbook on technological forecasting, as applied to the field of telecommunications.
I worked for a few years in telecommunications – first Western Electric, then TRW (neither of which exist today – not certain what that means). But I spent little time on forecasting – so I went back to graduate school – this time for a Ph.D. from Penn State University. I was lucky enough to have Keith Ord as my statistics professor – and it was my time in his classes that convinced me to do my dissertation on forecasting using technological growth curves, with Keith as my advisor.
And I’ve been studying time series data and creating forecasts ever since.