convore.json/groups/rstats/time-series/messages.json

[{"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1300559390.2097321, "message": "good [R]eferences for time series in the context of finance and econometrics: http://metaoptimize.com/qa/questions/749/introductory-booktext-on-time-series-analysis-and-forecasting", "group_id": 6150, "id": 389044}, {"user_id": 34017, "stars": [], "topic_id": 13979, "date_created": 1306165945.3819349, "message": "I do quite a bit of r analysis of financial and economic time series at http://timelyportfolio.blogspot.com.  I would love to engage in conversation about this topic.", "group_id": 6150, "id": 1154070}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1306170821.736527, "message": "@timelyportfolio some of your posts include \"eigenvalue ratio plots\" -- kindly tell us what they show and how they might be useful in constructing a portfolio.", "group_id": 6150, "id": 1154703}, {"user_id": 34017, "stars": [], "topic_id": 13979, "date_created": 1306268249.295346, "message": "Great question.  I hope to answer in a blog post later this week.  I will reply here once complete.", "group_id": 6150, "id": 1170173}, {"user_id": 34017, "stars": [], "topic_id": 13979, "date_created": 1306346894.0288219, "message": "I still plan to break the plot into its components on my blog, but http://www.google.com/url?sa=t&source=web&cd=6&sqi=2&ved=0CD0QFjAF&url=http%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fdownload%3Fdoi%3D10.1.1.19.1268%26rep%3Drep1%26type%3Dpdf&ei=AvfbTYTNIsajtgeYtNyvDw&usg=AFQjCNHA_5n6gfHnxeYtUvry-ACPMIanaw&sig2=kQGla3IdsSIIDjoQRlzG5Q pages 4 and 5 offer a great introduction to plot.  In addition to the components, I hope to be able to directly apply the methods to portfolio construction.", "group_id": 6150, "id": 1181726}, {"user_id": 34017, "stars": [{"date_created": 1306368052.2593811, "user_id": 10890}], "topic_id": 13979, "date_created": 1306356215.9008629, "message": "maybe this post http://timelyportfolio.blogspot.com/2011/05/eigen-who-how-can-i-write-about-eigen.html will help", "group_id": 6150, "id": 1184132}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1306367735.298368, "message": "@timelyportfolio thank you very much for your wonderful and illustrative post. You also gave an excellent reference paper by Friendly: http://www.math.yorku.ca/SCS/Papers/corrgram.pdf -- which discussses many interesting ways of visualizing a large correlation matrix.", "group_id": 6150, "id": 1185813}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1306368039.191349, "message": "following the pointers in the @timelyportfolio post, an \"eigenvector plot\"  plots each variable as a vector whose endpoint is the coordinates of the first two eigenvectors of the correlation matrix.", "group_id": 6150, "id": 1185860}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1306368608.3736441, "message": "@timelyportfolio looking forwarding to the second part of your answer. Perhaps it must go beyond eigenvector plots because an optimal portfolio not only takes into account the covariance matrix, but also the vector of returns. Easy to see though how such plots would be useful in real-time for statistical arbitrage. (Maybe this conversation should be moved over to https://convore.com/rstats/algorithmic-trading-using-r/)", "group_id": 6150, "id": 1185951}, {"user_id": 34017, "stars": [], "topic_id": 13979, "date_created": 1306426139.202239, "message": "just reading the presentations from R in Finance 2011 and found this very fine piece http://www.rinfinance.com/agenda/2011/StefanoIacus.pdf", "group_id": 6150, "id": 1193499}, {"user_id": 34017, "stars": [{"date_created": 1307163329.2222879, "user_id": 10890}], "topic_id": 13979, "date_created": 1307076016.9843061, "message": "another good source of information on eigenvalues http://math.nyu.edu/faculty/avellane/Lecture2Risk2010.pdf", "group_id": 6150, "id": 1274587}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1307163733.8483829, "message": "@timelyportfolio thank you very much for your pointer which essentially shows how  eigenvectors are useful in the replication of a portfolio in reduced dimension, e.g. the S&P500 index via a 5-factor model. Will you be posting something in connection with Random Matrix Theory?", "group_id": 6150, "id": 1284598}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1307286076.323462, "message": "@timelyportfolio are you also known as quantivity? Posted today, \"Covariance, Correlation, and RMT\" http://quantivity.wordpress.com/2011/06/05/covariance-correlation-and-random-matrix-theory/ which is really a bibliography for Random Matrix Theory in finance.", "group_id": 6150, "id": 1293556}, {"user_id": 844, "stars": [], "topic_id": 13979, "date_created": 1307311167.8355229, "message": "Posted a question on Stackoverflow, really basic on a time series issue and got introduced to the zoo library, wow.. something I did not realize was key to getting time series in R..", "group_id": 6150, "id": 1296298}, {"user_id": 844, "stars": [], "topic_id": 13979, "date_created": 1307311312.383847, "message": "Link to basic question and resources http://stackoverflow.com/questions/6244882/beginning-r-date-plots-and-analysis-with-matrix-object/", "group_id": 6150, "id": 1296345}, {"user_id": 34017, "stars": [], "topic_id": 13979, "date_created": 1307330124.8205459, "message": "Am not quantativity but do like his/her work.  Xts is a huge improvement to zoo.", "group_id": 6150, "id": 1299977}, {"user_id": 844, "stars": [], "topic_id": 13979, "date_created": 1307359079.4694259, "message": "Appreciate the note on Xts,, will investigate, I'm sort of surprised that I did not see any of these mentioned in the initial material I have been studying, is there a better resource?", "group_id": 6150, "id": 1302041}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1307412211.782794, "message": "@dartdog check out the chapter on time series in the \"R Cookbook\" -- on packages. And misc tips (esp. pertaining to your stackoverflow question: data formats and assigning types).", "group_id": 6150, "id": 1315865}, {"user_id": 844, "stars": [], "topic_id": 13979, "date_created": 1307416328.7425711, "message": "Is that avail on net? or only by purchase? got the r nutshell book but not the cookbook, trying to save $$ any net resource?", "group_id": 6150, "id": 1316363}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1307890588.0436881, "message": "review of financial applications of Random Matrix Theory http://docs.google.com/viewer?url=http%3A%2F%2Farxiv.org%2Fpdf%2F0910.1205v1", "group_id": 6150, "id": 1374491}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1308270666.7474761, "message": "But now suppose the correlations vary according to regimes... RollingCorrelation() function in the PerformanceAnalytics package: http://www.quora.com/Is-there-a-mathematical-way-to-measure-and-present-the-changing-correlations-between-two-stocks-in-different-market-conditions/answer/Nick-White-2", "group_id": 6150, "id": 1414971}, {"user_id": 34017, "stars": [], "topic_id": 13979, "date_created": 1308339513.2064281, "message": "much more comprehensive PCA on stocks http://blog.revolutionanalytics.com/2011/06/big-data-pca.html", "group_id": 6150, "id": 1421981}, {"user_id": 10890, "stars": [], "topic_id": 13979, "date_created": 1312132069.539865, "message": "for univariate time series, Measures of forecast accuracy http://ff.im/IRXzc -- cf. M-competition and M3-competition", "group_id": 6150, "id": 1759373}]