Abstract:

Online product reviews and related reviewers’ information are considered the most essential resource of electronic commerce. Although online reviews and third-party sites are perceived to be independent and unbiased, many studies have confirmed the existence of different types of biases in the product reviews. We use the Hodrick-Prescott Filter to estimate the bias in online third-party product reviews, and analyze the characteristics of the bias using ARMA model and impulse response function. Then discuss the influence factors of biases using movie panel data, proved the existence of sequential bias and which depends on.

Key words: Third-Party product reviews, Online reviews, Review bias, Sequential bias