Saturday, November 14, 2015

2015.11.13 VIX Cycle Model Chart

2015.11.13 VIX Cycle Model Chart

A reminder that all models are wrong but some are useful is appropriate at this time. While the model does not capture very short-term fluctuations, it has performed reasonably well in the long term, in spite of a continued bias of the phase shift into the future.  I have put the last two iterations of the model below to illustrate the mutability of the model over time, also shown on the blog here and here. Remembering that the fate of these models is they either adapt within their own "error" tolerance, or they fail to converge, or they converge but the predictive curve continues to overestimate or underestimate the price curve for a prolonged period.  The latter is what happened to the ill-fated DJIA model vs. 1.00!

2015.10.16 VIX Cycle Model Chart

2015.07.10 VIX Cycle Model Chart


Anonymous said...

Can you post the DJIA model chart so we can see what a failure of a model looks like? Since I expect the DJIA to fall again shortly I think it might be interesting to keep running it and see if it comes back into line. Wouldnt the model come back into line if the market sold off precipitously into year end? Also if you would what does this mean.."vs. 1.00"? Does that mean vs. reality?

Paolo said...

The old DJIA model is available on the blog, as is the new one.

If you expect any of these models to be "right" vs. reality, simply recall that suffering comes from expectation.