October 24, 2007

Bringing Rigor to Financial Forecasting

By adding quantitative discipline to their number crunching, financial analysts may avoid the psychological biases that can lead to judgment errors.


The old adage that there are three types of lies — lies, damned lies and statistics — could easily apply to accounting, because shifting numbers around can yield radically different results.

That’s why a group of analysts known as fundamentalists — for their bottom-up, microscopic approach to collecting and interpreting data — may benefit by stepping back from the minutiae and applying some quantitative rigor to their financial statement analysis (FSA), argues Professor Kenton Yee. FSA is the process of transforming accounting information from quarterly and annual statements into forecasts of future cash flow and price targets.

In contrast to the fundamentalists, quantitative financial analysts generally focus on the big picture by using computer modeling. Guided by statistical asset-pricing theories, quantitative analysts may purposely ignore details and hone in on one or two fundamental variables, like a firm’s earnings quality score. (Quantitative funds were disproportionately punished in the recent market sell-off, with critics panning their aggressive positions, admits Yee, but he says their troubles are not caused by how they process accounting information.)

“In fundamental analysis, changing a few parameters generates a whole different outcome, which can lead to irrational decisions,” Yee explains. “Fundamentalists can improve their performance by supplementing their analyses with quantitative processes and frameworks that instill discipline and prevent errors caused by psychological tendencies to misinterpret details.”

Though the fundamentalists’ in-depth approach often adds value because field research can uncover hidden problems, too much data can cloud the big picture, inviting analysts to identify patterns — real or imagined — and draw unfounded conclusions from those patterns, Yee says. Abundant data, in turn, might push analysts toward making irrational decisions based on irrelevant, fleeting or anecdotal information, he adds.

Moreover, cleaning up or tweaking accounting numbers to correct for bookkeeping anomalies and earnings manipulation is also prone to psychological pitfalls, he points out, because the many subjective choices required can lead to irrational decision making.

In his new research, Yee identifies 10 psychological pitfalls fundamentalists are subject to, including overconfidence, the belief that a favorable outcome is more likely than what will actually occur; overoptimism, overestimating the role of chance or skill, rather than making decisions based on realistic probabilities; and overreaction to recent events, such as a big announcement by a prominent company. Other judgment biases common to fundamentalists include hindsight bias, an overconfidence that an anecdotal scenario will reoccur; herding, following mass movements irrespective of logic; and myopia, a focus on recent, and perhaps exceptional, changes.

These behavioral biases can influence which nuggets of information an analyst focuses on or ignores, the frequency of forecast revisions and any subsequent conclusions, Yee notes.

“The emphasis instead should be on a consistent analytical process,” Yee says. “Quantitative measures such as quantified scorecards and portfolio constraints can add value by disciplining analysts to resist their own subjectivity. The key is committing to an intellectual paradigm you believe in and being disciplined enough to stick with it in the long haul irrespective of short-term results.”

This idea that committing to a disciplined FSA process can keep fundamentalists from interjecting psychology and emotions into their analyses has roots, Yee says, in a 2,500-year-old concept: Plato’s rule of law. The rule of law states that government authorities can exercise power only within boundaries delineated in written, publicly disclosed laws, which constrain police and prosecutors from behaving in capricious or self-serving ways. When applied to financial analysis, the principle suggests that a similar discipline will prevent fundamental analysts from acting on emotion.

“Ideally, a conclusion from FSA should not depend on the personality or passion of the analyst performing the exercise,” Yee says. “A disciplined process adds value by forcing analysts to make emotionally detached assessments that conform to preagreed stock-selection criteria.”

Kenton Yee is assistant professor of accounting at Columbia Business School.

Have You Read Columbia Business?

Columbia Business School’s alumni magazine connects alumni with each other and the School; celebrates alumni milestones and accomplishments; and chronicles the impact of Columbia Business School alumni, faculty members, and students on the global business landscape.

Read Columbia Business>