It appears that the US stock market’s rally this year has trashed the bear-market calls made at the end of 2018 and repeated in late-January, based on a version of the Hidden Markov model (HMM). But a fresh run of The Capital Spectator’s application of HMM for profiling the big-picture trend is doubling down on the analysis of the S&P 500 Index. Animal spirits have revived in 2019 after a sharp slide late last year, but the model continues to insist that a bearish bias endures. Taking the analysis at face value, if you’re so inclined, suggests that this year’s bounce is a bear-market rally.
The model is currently estimating a 96% probability that the S&P 500 remains in a bear-market state (as of the S&P’s March 7 close). For perspective, the estimate rose above 50% for the first time in several years in late-December. By that standard, this bear-market call will likely endure for the foreseeable future.
As a refresher on the design, the analytical engine focuses on rolling one-year returns for the S&P via two profiles. The first is a slower but somewhat more reliable framework of using average monthly data (updated daily for the current month). The second review crunches daily numbers, which react faster to trend changes but suffer a greater degree of noise. Taking the average of the two on a daily basis seeks to find the sweet spot between timeliness and reliability.
Reliability, of course, is always south of 100% for this model — and anything else you can dream up in the cause of forecasting the stock market’s trend. But this twist on HMM has an encouraging history (for some background, see these posts here and here). Can you rely on it as the one and only last word on modeling the market? No, but it deserves a role as one of several analytical tools for evaluating the near-term outlook and on that point it’s safe to assume that this model continues to make a case for a cautious stance for US stocks.
Yes, many analysts and investors disagree and there’s no way to know for sure if they’re right… or wrong. In the long run, it’s prudent to assume that the market will generate a respectable performance – close to 6% a year, plus whatever you’re expecting for the risk-free rate, according to one model.
In the near term, by contrast, the outlook is considerably less sanguine, or so the HMM-based analysis continues to advise. Eye-balling a price chart that tracks the recent history of the S&P doesn’t offer an obvious counterpoint, at least not yet.
Granted, the Capital Spectator’s HMM profiling is just one model and nothing’s perfect. But in the grand scheme of evaluating risk factors, HMM offers a reason to remain wary of the S&P 500’s near-term prospects. That puts the onus on the bulls to come up with analytics to argue otherwise.
Learn To Use R For Portfolio Analysis
Quantitative Investment Portfolio Analytics In R:
An Introduction To R For Modeling Portfolio Risk and Return
By James Picerno