Investors nowadays post heterogeneous sentiments on social media about financial assets based on their trading preferences. However, existing works typically analyze the sentiment by its content only and do not account for investor profiles and trading preferences in different types of assets. This paper explicitly considers how investor sentiment about financial market events is shaped by the relative discussions of different types of investors. We leverage a large-scale financial social media dataset and employ a structural topic modeling approach to extract topical contents of investor sentiment across multiple finance-specific factors. The identified topics reveal important events related to the financial market and show strong heterogeneity in the social media content in terms of compositions of investor profiles, asset categories, and bullish/bearish sentiment. Results show that investors with different profiles and trading preferences tend to discuss financial markets with heterogeneous beliefs, leading to divergent opinions about those events regarding the topic prevalence and proportion. Moreover, our findings may shed light on the mechanism that underlies the efficient investor sentiment extraction and aggregation while considering the heterogeneity of investor sentiment across different dimensions.

No abstract

Estimations and evaluations of the main patterns of time series data in groups benefit large amounts of applications in various fields. Different from the classical auto-correlation time series analysis and the modern neural networks techniques, in this paper we propose a combination of functional analysis of variance (FANOVA) and permutation tests in a more intuitive manner for a limited sample size. First, FANOVA is applied in order to separate the common information and to dig out the additional categorical influence through paired group comparison, the results of which are secondly analyzed through permutation tests to identify the time zones where the means of the different groups differ significantly. Normalized kernel functions of different groups are able to reflect remarkable mean characteristics in grouped unities, also meaningful for deeper interpretation and group-wise classification. In order to learn whether and when the proposed method of FANOVA and permutation F-test works precisely and efficiently, we compare the estimated kernel results with the ground truth on simulated data. After the confirmation of the model's efficiency from simulation, we apply it also to the RAVDESS facial dataset to extract the emotional behaviors of humans based on facial muscles contractions (so-called action units (AU) technically in computer graphics), by comparing the neutral performances with emotional ones.