|
|
FlockingMembers Members of the Finance-Flocking Group
Papers Should Network Structure Matter in Agent-Based Finance? S. Alfarano and M. Milakovic Flocks, Herds, and Schools: A Distributed Behavior Model. C.W. Reynolds A Simple Model of Global Cascades on Random Networks. D. J. Watts Novel Type of Phase Transition in a System of Self-Driven Particles. T. Vicsek et. al. Tamas Vicsek 1995 Paper on Flocking Empirical properties of asset returns: stylized facts and statistical issues Maybe this can help us look for features and impact co-efficients in news items. This also mentions the cyclic effects of news on stocks and stocks on news: Information sources, news, and rumors in financial markets: Insights into the foreign exchange market Other stuff Something interesting: Check out the links to Recognia on the page. They manually look for and announce about 50 patterns in market behavior. Chart game Model Here is a pdf for the formulas. Attach:model.pdf Code iGraph is an R/Python package that we can use to generate and display different types of social networks. Here is the link: http://cneurocvs.rmki.kfki.hu/igraph/documentation.html MATLAB Code for the BA algorithm http://webscripts.softpedia.com/script/Scientific-Engineering-Ruby/Statistics-and-Probability/B-A-ScaleFree-Network-Generation-and-Visualization-35557.html Plots Plot of kernel density estimate for the mean estimate of the flock after 50 steps versus a kernel density estimate for the 5000 samples from the corresponding normal distribution.  Plot of the mean estimate for the flock over 50 time steps  Histogram of the joint (r(t+day,day),r(t,day)). A histogram of the number of returns at time t+day between (a,b] and the return was between (c,d] at time t. The ranges for the histogram was determined by finding percentiles of the data 0:1:100  Same as above except delta t = 5 min  Same as above except delta t = 30 min  |