Stock Data Analysis Functions (lcyanalysis)
Analysis of stock data ups and downs trend, the stock technical analysis indicators function have trend line, reversal pattern and market trend.

Subsampling Winner Algorithm for Classification (swa)
This algorithm conducts variable selection in the classification setting. It repeatedly subsamples variables and runs linear discriminant analysis (LDA) on the subsampled variables. Variables are scored based on the AUC and the t-statistics. Variables then enter a competition and the semi-finalist variables will be evaluated in a final round of LDA classification. The algorithm then outputs a list of variable selected. Qiao, Sun and Fan (2017) <http://…/swa.html>.

Computation of the Orbit-Aware Quad Census (oaqc)
Implements the efficient algorithm by Ortmann and Brandes (2017) <doi:10.1007/s41109-017-0027-2> to compute the orbit-aware frequency distribution of induced and non-induced quads, i.e. subgraphs of size four. Given an edge matrix, data frame, or a graph object (e.g., ‘igraph’), the orbit-aware counts are computed respective each of the edges and nodes.

Optimal Timing Identification (OptimalTiming)
Identify the optimal timing for new treatment initiation during multiple state disease transition, including multistate model fitting, simulation of mean residual lifetime for a given transition state, and estimation of confidence interval. The method is referred to de Wreede, L., Fiocco, M., & Putter, H. (2011) <doi:10.18637/jss.v038.i07>.

New and Extended Plots, Methods, and Panel Functions for ‘lattice’ (tactile)
Extensions to ‘lattice’, providing new high-level functions, methods for existing functions, panel functions, and a theme.

An Integrated Framework for Textual Sentiment Time Series Aggregation and Prediction (sentometrics)
Time series analysis based on textual sentiment, accounting for the intrinsic challenge that sentiment can be computed and pooled across texts and time in many ways. As described in Ardia et al. (2017) <>, the package provides a means to model the impact of sentiment in texts on a target variable, by first computing a wide range of textual sentiment measures and then selecting those that are most informative.