The Lead-Lag Puzzle of Demand and Distribution: A Graphical Method Applied to Movies

Abstract

Understanding the lead-lag relationship between distribution and demand is an important and challenging issue for all marketers. It is particularly challenging in the movie industry, where the very short lifespan and decaying revenue and exhibition patterns of motion pictures means that the associated time series are short and nonstationary, rendering existing econometric methods unreliable. We propose an alternate method that uses state-space diagrams to determine lead-lag relationships. Straightforward to apply and interpret, it takes advantage of the eye’s ability to see patterns that algebra-based formulations cannot easily recognize. A number of validation tests are provided to illustrate the usefulness and limitations of the method. We study the weekly data for 231 major movies released in 2000–2001. While econometric methods do not provide consistent results, the graphical method of visually inferred causality clearly shows a pattern that demand leads distribution for most movies. In other words, the dominant industry pattern is one of movie exhibitors monitoring box office sales and then responding with screen allocation decisions. The managerial implications of these findings are discussed.distribution, marketing tools, movies, state-space diagrams, time series

Finite Mixtures of Generalized Linear Regression Models

“…The box office and theaters data for 407 movies playing between May 5, 2000 and December 7, 2001 were collected from a popular website of movie records (www.the-number.com), see Krider et al (2005). The gross box-office takings for the 40 most popular movies for each weekend in the time period are recorded and transformed into market shares to account for the difference in volume between weekends.…”

Demand and distribution relationships in the ready-to-drink iced tea market: A graphical approach

“…If they are perfectly out of phase (i.e., phase difference is 180°), corresponding to negatively correlated contemporaneous effects, the plot will collapse to a straight line with a negative slope. As discussed in Krider et al (2005), state-space diagrams offer the useful feature that curvature suggests a lead-lag relation and which series leads. Their main advantages are that if there is a lead-lag relationship, the curvature can be quickly detected by eye, and that changes in the direction as the market evolves can be easily determined.…”

“…In this research, we show how a recently introduced simple graphical method (Krider et al 2005) can be used for initial explorations of two coupled time series. The method is based on state-space diagrams.…”

“…The trajectory of the system in the state-space is typically shown by joining the plotted states with a line in the order of increasing time. By plotting state-space diagrams for two related time series and observing the patterns of the trajectory, Krider et al (2005) discuss how this graphical method takes advantage of the eye's ability to see dynamic patterns that algebra based formulations cannot easily recognize. The state-space plots show not only evolving trends, but also lead-lag relationships between the variables.…”

“…The use of state-space diagrams in Krider et al (2005) was motivated by the desire to untangle the causal relationship between demand and distribution for very short nonstationary time series, with which econometric methods may have difficulty. In an application to motion pictures, they show that the method provides more robust results than a VAR model with the Granger causality test.…”