Structural equation modeling for objective financial forecasts
I'm curious as to what drives your predictions. Whilst it is easy to see equities plummeting, it is hard to envisage gold going back below $800 an ounce, but that is what your long-term chart appears to show. Are you able to enlighten me please? Many thanks.
The predictions are driven by the cycle parameters which are in turn mathematically determined from the time series. So, for example, in the case of gold, looking at the largest amplitude cycle - which is also the longest - the full cycle is 13,130 ± 12 days, which is ± standard error. The phase is -1567 days ± 30 days, and the relative amplitude is -0.784 ± 0.001. If you are aware of how structural models work, those are very low standard errors for parameters indicating a low probability (p<0.001) that the parameters fit the data by chance.There are a number of such cycles operating in concert, with lower magnitude periods, for example, the next one is 8,708 days ± 21 days, etc.But the map, or model, is not the territory, and in this case, the territory is a dynamic time series, not a static topo map. With the added disadvantage that the measurement unit, the dollar in this case, is itself a dynamically changing metric.One way of looking at it is that if indeed gold prices follow a cyclic pattern, then there is a strong attractor for lower prices ahead. On the other hand, reality may throw a few curve balls between now and then. Like there ain't no such thing as a free lunch, there ain't no such thing as a perfect structural model!
Post a Comment