A Forest of Worlds
99/
Mortgage default probabilities, asset price changes, macroeconomic variables like GDP and inflation could all be reduced to immutable averages and variances with a little data “munging,” “scrubbing,” or “cleaning.” All verbs that call to mind the image of Cinderella wiping her sister’s filth off of the bathroom floor. That was my job. It wasn’t supposed to be my job. Actually, it was supposed to be beneath even me. But I was the face of the product and if the numbers were wrong then I was the face of the wrong numbers as well.
I do not like being the face of wrong numbers.
Once the statistical information had been prepared the quants could calculate precise values for assets and optimal weightings for portfolios. And they could do it in an ordinary spreadsheet without a lot of fancy software.
It was all because of this miracle that the quants had pulled off, which was to get rid of time (and therefore change) from their formulas. I suddenly understood why I couldn’t get anything done in Excel. I hadn’t learned the quant skill of banishing time from my mind. I was imagining evaluating millions of permutations with my models. A vast forest of possible financial worlds.
But here I was again thinking like a stupid computer programmer that just couldn’t understand the power of statistics. You see, the quants didn’t need to evaluate a whole forest of possible worlds because they were convinced it wouldn’t matter in the end.
Why were they so sure that the statistical measures were stable? I have no idea, go ask a quant. But I had a feeling they were wrong. To prove it, I only had to explore this magical forest of possibilities and bring back the evidence.
For that, all I needed was a software product like a spreadsheet in some ways, but much more powerful. It felt like my program was just around the corner from being used in this fashion.
If I just kept hacking I would get there.
