CMStatistics 2022: Start Registration
View Submission - CMStatistics
B0588
Title: Exploring pre-launch movie electronic word-of-mouth time series by functional data analysis Authors:  Tianyu Guan - Brock University (Canada) [presenting]
Abstract: the aim is to explore the dynamic patterns of the pre-launch online movie reviews, or movie electronic word-of-mouth (eWOM), over time and investigate the impact of pre-launch eWOM on explaining the box office revenues. One focus is to use the pre-launch eWOM evolution data as an early indicator of strong or weak box office, which would be helpful to business decision-makers in the movie industry. The eWOM data evolve in time and are treated as functional data. We observe that most eWOM data exhibit a positivity bias and extremity; therefore, we apply the functional principal component analysis, a dimension reduction technique in functional data analysis, to explore the dynamic patterns of various quantile trajectories of the movie eWOM, instead of directly studying the eWOM trajectories. The functional principal component (FPC) scores of quantile trajectories at various quantile levels are used to explain the box office revenues. We use the sparse group lasso method to select the quantile levels and individual FPC scores that make significant contributions to the prediction of box office revenues. The results show that compared with other measures such as valence and variance, the top-end quantiles would be a better measure in capturing the relations between the pre-launch product ratings time pattern and launch sales.