Predicting stock prices is possible

Predicting stock prices is possible, according to a new study from the Tippie College of Business at the University of Iowa, especially during short windows.

The authors suggest that up to one minute after a stock leaves the confines of its bid-ask spread, price movements can be predicted with better than 50-50 accuracy rate.

The predictability window, which can last up to a maximum of thirty minutes, is strongest during the first five minutes after a stock leaves the spread, the researchers inform.

Researchers Michael Rechenthin and Nick Street say their findings challenge the generally held belief that it is not possible to predict stock prices.

Currently, most people believe that while financial reports and news can move stock prices, there is nothing inherent in the trend of a price that can be used to forecast its next movement.

Their study, “Using conditional probability to identify trends in intra-day high-frequency equity pricing,” has been published in the journal Physica A.

Rechenthin, a former Chicago Stock Exchange floor trader, said:

“This study is the first step in showing that there is predictability, and that once a price escapes the confines of the bid-ask spread, it’s showing a trend.”

“In other words, it’s more than just a coin flip where the price goes.”

Predicting stock prices during the bid-ask spread not possible

The study looks at building models for predicting the direction of future stock price. It examined the price movements of the S&P 500 exchange traded stock fund during 2005, which holds all 500 S&P’s stocks and is considered representative of the overall American market.

According to their analysis, it is not possible to predict a stock’s price within the bid-ask spread – i.e. the space between the price that purchasers are willing to pay for a stock (bid) and the price vendors are willing to sell it for (the ask) – as the market tries to establish the value of an asset.

However, as soon as the traders have set a value and the price escapes that spread, predicting becomes possible. The researchers tracked the stock’s price at 1, 3, 5, and 20 seconds and 1, 5 and 30 minutes.

Stocks more likely to reverse recent trends

In most cases the stock price breaks the spread after five to ten seconds, the authors found, “and the predictability of its subsequent movements depended on the pattern of its most recent trades.”

For example, if during the stock’s two most recent trades it had moved up and then down, there was a 52% chance that the trend would reverse itself within 5 seconds, and 43% probability within 20 seconds.

According to Rechenthin, only previous trade prices can drive these trends, because “other factors that influence price, such as financial statements or news, cannot be incorporated into a price in such a short window.”

People may wonder how a 52% probability helps, as it is so close to a normal 50% probability. Street explains that in stock trading this is a significant increase, “and something that can be exploited.”

The researchers said they now plan to develop a working model that exploits these probabilities for more efficient trading.