JPMorgan researchers have come up with a relatively simple way to predict what’s next for the stock market, and it has good news for investors. Strategist Nikolaos Panigirtzoglou said in a note to clients that his team updated a stock prediction model that currently sees the market moving higher in the back half of the year. The model uses six signals — economic momentum, price momentum, turnover, value, positions and flows — to predict the move in the S & P 500 over the next six months. The latest reading is quite confident that stocks will go up. “The current probability of 96% is well above the 75% cutoff, implying low probability of a down move in the S & P 500 over the next six months based on the model inputs,” Panigirtzoglou wrote. .SPX YTD mountain The S & P 500 was flirting with a new record high on Thursday. Of the signals used in the model, flows are currently the most supportive of the idea that stocks will rise. That signal is based on the difference between equity and bond fund flows tracked by the firm, relative to history. That data, published in the same note, shows that equity funds have seen outflows over the past four weeks, while bond funds have seen inflows. The same dynamic holds when looking at just U.S. funds. This means the S & P 500 has trended back up to near record levels even with many investors whittling down their stock exposure. Even retail traders are not all-in on stocks at the moment, according to JPMorgan. “The retail impulse into equities improved somewhat in June after a very weak May but it still looks modest vs. its recent history,” Panigirtzoglou wrote. The latest update is a change from how the JPMorgan model used to work. The strategist found that putting those six variables into a logit model — a regression calculation that is simple by Wall Street standards — performed better than using an artificial intelligence model with similar inputs, as well as several other modeling approaches. In particular, the logit model excelled at predicting down moves in the market relative to other techniques. — CNBC’s Michael Bloom contributed reporting.
