4 ms·

IDK if you'd call it a strategy, but I'm starting to chain some code together so I can apply technical indicators to various climate indices. It's a fun science

IDK if you'd call it a strategy, but I'm starting to chain some code together so I can apply technical indicators to various climate indices. It's a fun science project if nothing else. But pragmatically speaking it helps me to apply the technical indicators I'm familiar with to the data that's making the news, and form some of my own perceptions of what's going on. The most extreme direct risks to me from climate affecting weather seem to be related to wildfires, but the local response (fire, police, mental illness-related efforts, various non-profits) has really raised its game after recent fires, and now with PG&E making some changes things don't seem as bad.

noduerme 4y ago
Is one of the goals to apply climate indices to economic predictions? I've been thinking a lot recently about a chart I saw a long time ago, showing how "peak oil" (as it was called before fracking increased reserves) would result in a series of economic shocks where the global economy bounced up against an energy ceiling, sending energy prices higher until the economy entered recession, at which point energy demand/prices would fall enough to trigger another growth boom that would again be cut short by the oil ceiling. This presupposed that oil availability was falling every year after the peak. But it also supposed that energy was the only "ceiling" out there. Climate change, covid, the size of the generation entering old age, market bubbles, and trailing energy capacity have all served in some way to compress and limit growth in the last ten years. I'm curious what kind of proxies you're using to model the interactions.
- themodelplumber 4y ago
  For now it's a standalone analysis (constrained to climate) looking back & forward, applying a variety of indicators, some normal stuff like mas & stddev, some shamelessly ripped from equities trading books (bollinger bands as an example building on those first items), along with some of my own which are probably dumb but which seem reliable enough for me. I admit I'm not a huge fan of predictive or causal models unless taken in the aggregate, usually. Especially the singular-cause or singular-outcome kind. This is coming from personal interest in this topic within psychology study...for one I find that such models can become pretty breathless & kind of suck all the energy out of the room, energy that can be redirected into an executive stance. Where otherwise you end up with perfectly good mental resources stuck in a perceptive stance. It's cool to predict, but also even cooler to predict with an open mind to many possible outcomes, and then also even cooler to do that and then say "now here's the plan" iow.

noduerme 4y ago

Is one of the goals to apply climate indices to economic predictions? I've been thinking a lot recently about a chart I saw a long time ago, showing how "peak oil" (as it was called before fracking increased reserves) would result in a series of economic shocks where the global economy bounced up against an energy ceiling, sending energy prices higher until the economy entered recession, at which point energy demand/prices would fall enough to trigger another growth boom that would again be cut short by the oil ceiling. This presupposed that oil availability was falling every year after the peak. But it also supposed that energy was the only "ceiling" out there. Climate change, covid, the size of the generation entering old age, market bubbles, and trailing energy capacity have all served in some way to compress and limit growth in the last ten years. I'm curious what kind of proxies you're using to model the interactions.

themodelplumber 4y ago
For now it's a standalone analysis (constrained to climate) looking back & forward, applying a variety of indicators, some normal stuff like mas & stddev, some shamelessly ripped from equities trading books (bollinger bands as an example building on those first items), along with some of my own which are probably dumb but which seem reliable enough for me. I admit I'm not a huge fan of predictive or causal models unless taken in the aggregate, usually. Especially the singular-cause or singular-outcome kind. This is coming from personal interest in this topic within psychology study...for one I find that such models can become pretty breathless & kind of suck all the energy out of the room, energy that can be redirected into an executive stance. Where otherwise you end up with perfectly good mental resources stuck in a perceptive stance. It's cool to predict, but also even cooler to predict with an open mind to many possible outcomes, and then also even cooler to do that and then say "now here's the plan" iow.