This paper proposes new regressive proportional and partial proportional odds models and a framework to predict trajectories of repeated ordinal outcomes, which is a new development. We illustrated the proposed models...
详细信息
This paper proposes new regressive proportional and partial proportional odds models and a framework to predict trajectories of repeated ordinal outcomes, which is a new development. We illustrated the proposed models using repeated ordinal responses on activities of daily living from older adults collected biannually through the Health and Retirement Study in the USA. The proposed framework uses the marginal and conditional modeling approach to obtain the joint model and predict the joint probability of a sequence of ordinal outcomes and trajectories. Besides, these models significantly reduce over-parameterization, as one needs to fit one model for each follow-up. This model allows assessing the effect of prior responses on current outcomes, including interaction terms among previous responses and between prior outcomes and covariates in the model. Also, it permits the varying number of risk factors for each follow-up. The prediction accuracy for full, training, and test data is close and varies between 0.91 and 0.94. The bootstrap simulation demonstrates the bias of parameter estimates, accuracy, and predicted joint probabilities are negligible with very low mean squared error. This model and framework would be instrumental in studying trajectories generated from longitudinal studies. The proposed framework can be used to analyze bigdata generated from repeated measures. This model readily uses a divide and recombine approach for bigdata in a statistically valid manner.
We present BiDaML 2.0, an integrated suite of visual languages and supporting tool to help multidisciplinary teams with the design of bigdata analytics solutions. BiDaML tool support provides a platform for efficient...
详细信息
We present BiDaML 2.0, an integrated suite of visual languages and supporting tool to help multidisciplinary teams with the design of bigdata analytics solutions. BiDaML tool support provides a platform for efficiently producing BiDaML diagrams and facilitating their design, creation, report and code generation. We evaluated BiDaML using two types of evaluations, a theoretical analysis using the "physics of notations", and an empirical study with 1) a group of 12 target end-users and 2) five individual end-users. Participants mostly agreed that BiDaML was straightforward to understand/learn, and prefer BiDaML for supporting complex data analytics solution modeling than other modeling languages.
We perform a neural network analysis of the impact of Russian retail investors ' sentiment on the stock price behavior of well-known American companies. We study American stocks in a situation of a time-segmentati...
详细信息
We perform a neural network analysis of the impact of Russian retail investors ' sentiment on the stock price behavior of well-known American companies. We study American stocks in a situation of a time-segmentation of the stock market. A special feature of our analysis is the separate time trading mode, when trading is active at the SPB (formerly St. Petersburg) exchange and inactive at the US stock exchanges. Building on the unique local exchange data and original technique for constructing a neural network to identify the sentiment of messages from several Internet forums, we uncover the existence of behavioral anomalies in a non-English-speaking emerging market and analyze sentiment and attention metrics in social networks. We construct several sentiment metrics based on AI text analysis and use panel regression to identify their statistical significance under the selected hypotheses. The impact of sentiment is examined across the entire sample of US companies available to investors on the SPB exchange and a separate zooming is made at the top 10, 25, 50, and 100 stocks that are under special interest manifested by volume of discussions and trading volume. We also analyze the impact of sentiment on price reaction for individual popular stocks and by industry. We find that retail investors' sentiment exercises a statistically significant influence on price spikes. The stocks, most sensitive to sentiment, are healthcare and high tech.
暂无评论