Working Knowledge
Business Research for Business Leaders
Filter Results: (23) Arrow Down

Forecasting and PredictionRemove Forecasting and Prediction →

New research on forecasting and predication from Harvard Business School faculty on issues including forecasting success or failure of new ventures, why countries have difficulty controlling budget deficits, and new models for predicting the success of marketing investments.
Page 1 of 23 Results →

Financial Meltdowns Are More Predictable Than We Thought

by Danielle Kost

Robin Greenwood and Samuel G. Hanson discuss new research that shows economic crises follow predictable patterns. Open for comment; 0 Comments.

Is the 'Experimentation Organization' Becoming the Competitive Gold Standard?

by James Heskett

SUMMING UP: Digital experimentation is gaining momentum as an everyday habit in many organizations, especially those in high tech, say James Heskett's readers. Open for comment; 0 Comments.

Business Reopening Decisions and Demand Forecasts During the COVID-19 Pandemic

by Dylan Balla-Elliott, Zoë B. Cullen, Edward L. Glaeser, Michael Luca, and Christopher Stanton

Findings from a nationwide survey underscore the importance of demand projections and interdependencies among businesses for owners’ reopening decisions. Businesses expect the demand for their services will be greatly depressed for many months to come.

Predictions, Prophets, and Restarting Your Business

by Frank V. Cespedes

Businesses are starting to plan their re-entry into the market, but how do they know what that market will look like? Frank V. Cespedes warns against putting too much trust in forecasters. Open for comment; 0 Comments.

Creating the Experimentation Organization

by Michael Blanding

New tools allow companies to innovate on an unprecedented scale, in every aspect of business. But the organization must also change. Stefan Thomke previews his forthcoming book, Experimentation Works. Open for comment; 0 Comments.

Don't Turn Your Marketing Function Over to AI Just Yet

by Kristen Senz

Lacking human insight, artificial intelligence will be limited when it comes to helping marketers open the black box of market prediction, says Tomomichi Amano. Open for comment; 0 Comments.

Use Artificial Intelligence to Set Sales Targets That Motivate

by Michael Blanding

Setting sales targets has always been an inexact science, with serious consequences if done poorly. Using AI-based advanced analytics might be the answer, argues Doug Chung. Open for comment; 0 Comments.

Using Models to Persuade

by Joshua Schwartzstein and Adi Sunderam

“Model persuasion” happens when would-be persuaders offer receivers a streamlined way of understanding data they already know, especially when the data is open to interpretation. Using examples from finance, politics, and law, the authors find that truthtellers do not eliminate the impact of misleading persuasion because wrong models may better fit the past than correct models.

Reflexivity in Credit Markets

by Robin Greenwood, Samuel G. Hanson, and Lawrence J. Jin

Investors’ biases and market outcomes affect each other in a two-way feedback loop. This study develops a model of a credit market feedback loop, finding that when investors become more bullish this can predict positive returns in the short run, even if expected returns become more negative at longer horizons.

Consumer Inertia and Market Power

by Alexander MacKay and Marc Remer

Consumers are often more likely to buy a product if they have purchased it previously. This paper provides a means to estimate the magnitude of this phenomenon (i.e., consumer inertia) and shows how it affects the prices of firms in competitive settings. Perhaps surprisingly, greater consumer inertia can result in smaller price increases after a merger.

Judgment Aggregation in Creative Production: Evidence from the Movie Industry

by Hong Luo, Jeffrey T. Macher, and Michael Wahlen

Selecting early-stage ideas in creative industries is challenging because consumer taste is hard to predict and the quantity to sift through is large. Using The Black List that ranks scripts annually based on nominations from film executives, this study shows that aggregating expert opinions helps reduce quality uncertainty and can influence high-budget production.

The Better Way to Forecast the Future

by Roberta Holland

We can forecast hurricane paths with great certainty, yet many businesses can't predict a supply chain snafu just around the corner. Yael Grushka-Cockayne says crowdsourcing can help. Open for comment; 0 Comments.

Demand Estimation in Models of Imperfect Competition

by Alexander MacKay and Nathan H. Miller

The study shows how knowledge about firm behavior can be modeled to better predict demand. Firms tend to raise prices in response to higher demand, so observed relationships between price and quantity can be quite misleading. The authors provide an adjustment that can be used when price experiments or instrumental variables are not available.

Forecasting Airport Transfer Passenger Flow Using Real-Time Data and Machine Learning

by Xiaojia Guo, Yael Grushka-Cockayne, and Bert De Reyck

Passengers arriving at international hubs often endure delays, especially at immigration and security. This study of London’s Heathrow Airport develops a system to provide real-time information about transfer passengers’ journeys through the airport to better serve passengers, airlines, and their employees. It shows how advanced machine learning could be accessible to managers.

Quantile Forecasts of Product Life Cycles Using Exponential Smoothing

by Xiaojia Guo, Kenneth C. Lichtendahl Jr., and Yael Grushka-Cockayne

Many important business decisions rely on a manager’s forecast of a product or service’s life cycle. One of the most widely used forecasting techniques is exponential smoothing models. This paper introduces a model suitable for large-scale forecasting environments where key operational decisions depend on quantile forecasts.

Nowcasting Gentrification: Using Yelp Data to Quantify Neighborhood Change

by Edward L. Glaeser, Hyunjin Kim, and Michael Luca

This study finds that data from digital platforms (in this case, Yelp) can help forecast which neighborhoods are gentrifying and provide new ways to measure business landscape changes that accompany demographic changes.

Nowcasting the Local Economy: Using Yelp Data to Measure Economic Activity at Scale

by Edward L. Glaeser, Hyunjin Kim, and Michael Luca

Data from online platforms ranging from Yelp to Zillow offer the potential for improved measurement of the local economy. This paper finds that Yelp data can predict business growth, as measured by the Census Bureau, before official statistics are released. Predictive power increases with population density, income, and education level.

Bubbles for Fama

by Robin Greenwood, Andrei Shleifer, and Yang You

Nobel laureate Eugene F. Fama has famously claimed that there is no such thing as a bubble, which he defines as a large price run-up that predictably crashes. Analyzing industry data for the US and internationally, the authors find that Fama is mostly right that a sharp price increase of an industry portfolio does not, on average, predict unusually low returns going forward. Yet the authors show that there is much more to a bubble than merely increases in prices; they show a number of characteristics that predict an end to the bubble.

Bias in Official Fiscal Forecasts: Can Private Forecasts Help?

by Jeffrey A. Frankel and Jesse Schreger

Why do countries find it so hard to get their deficits under control? The authors bring together data on private sector forecasts with official government forecasts of 26 countries, and identify systematic patterns in errors that official budget agencies make in their forecasts—particularly over-optimistic forecasts of GDP growth and budget balances. Incorporating private sector forecasts into the budget process could reduce violations of limits set by fiscal rules by identifying errors ahead of time rather than just afterwards.

How to Predict if a New Business Idea is Any Good

by Michael Blanding

Professor Pian Shu tackles one of the most difficult questions in the startup world: How can you tell if a new business will succeed? Open for comment; 0 Comments.