Thursday, December 26, 2019

"Text-reading machines can predict share prices"

Apparently not all bleepin' text readers.
On the other hand the authors cite an old friend, so there is that.

From the University of Chicago's Chicago Booth Review, December 18:
A single word in a news report—a well-placed “undervalue,” for example—can drive a company’s stock price up or down. Investors can benefit if they can figure out which words matter within a few days, research suggests.

Investors and researchers have suspected for decades that text could be used to predict markets, some trying and failing. But applying machine-learning techniques originated by computer scientists, Harvard’s Zheng Tracy Ke, Yale’s Bryan T. Kelly, and Chicago Booth’s Dacheng Xiu have built a model that in early tests outperformed a similar strategy based on scores from RavenPack, the leading vendor of news-sentiment scores.

Traditionally finance researchers and market practitioners have relied on accounting data and fundamentals to predict where the market is headed. But quarterly reports arrive slowly for a market moving at warp speed, which led researchers and traders to look for other sources of predictive information, including news. To find out if news reports could be used to predict stock prices, Ke, Kelly, and Xiu borrowed machine-learning techniques used by computer scientists, who are increasingly training machines to understand text.

Efforts to predict market direction by parsing financial journalism date back to 1933, when economist and businessman Alfred Cowles III classified pieces in the Wall Street Journal as bullishbearish, or neutral to inform trading strategies. That didn’t necessarily work—Cowles’s theoretical portfolio would have underperformed the market by more than 3 percent a year from 1902 to 1929, the researchers note—but other people have continued to pursue the idea of extracting useful information from text. Among them, Northwestern’s Scott R. Baker, Stanford’s Nicholas Bloom, and Chicago Booth’s Steven J. Davis analyzed years of newspaper articles to identify words associated with economic uncertainty, and have used those words to inform dozens of uncertainty-related indexes.   
Some efforts to assess sentiment in text rely on preexisting dictionaries created for other purposes—such as the Harvard-IV Dictionary, a manually selected list of positive and negative psychosocial words, and the Loughran-McDonald Master Dictionary, developed to highlight meaningful words in financial texts and the sentiment associated with those words. The latter starts with word lists and uses US Securities and Exchange Commission filings to add terms relevant to the finance sector. For example, the dictionary added Scholes for the Black-Scholes modeling tool used with financial derivatives.

Ke, Kelly, and Xiu created a model that essentially automatically generates a dictionary of relevant words and allows for contextually specific sentiment scores. Using supervised machine learning and a method that required only a laptop and basic statistical capabilities, the researchers analyzed more than 22 million articles published from 1989 to 2012 by Dow Jones Newswires. Classifying words as either positive or negative, the researchers generated article-level sentiment scores—to highlight how news likely to be perceived as positive or negative would impact stock prices....

We've been referring interested readers to Alfred Cowles III since January 2008, with the most recent occurrence being last month in "The Second Editor of The Wall Street Journal Was Pretty Good at Figuring Out Markets":
During a misspent youth Alfred Cowles was one of my heroes, with the Cowles Commission Monograph no. 3 being one of the great early statistical works on the stock market: "The Cowles Commission's Common Stock Indexes 1871-1937."*

The Cowles Commission has, over the years employed some bemedaled worker bees:
Tjalling Koopmans, Kenneth Arrow, Gerard Debreu, James Tobin, Franco Modigliani, Herbert Simon, Lawrence Klein, Trygve Haavelmo and Harry Markowitz, all of whom went on to win Nobel econ prizes.
The Commission is now at Yale where another Laureate, Robert Shiller, oversees the renamed "Cowles Foundation for Research in Economics"

In addition to Common Stock Indexes  we've posted some of Cowles' other stuff including a quick hit in January 2008, after the August 'ought-seven' quant-quake but before things got really ugly in September '08:

"Can Stock Market Forecasters Forecast?" is the title of a paper by Alfred Cowles III.

It appeared in Vol.1, No. 3 of Econometrica, after having been read to a joint meeting of the Econometric Society and the American Statistical Society.

Mr. Cowles answer to the question?
"It is doubtful."
December 31, 1932.
I go back and forth on the efficacy of market forecasts so we try to give the reader informed opinion and leave the decision-making to them.
It appears the question is bounded by the age-old trader's lament:

"As soon as you think you've found the key 
they go and change the lock"