tag:blogger.com,1999:blog-59909462781239249792024-03-19T05:17:35.451-07:00Xi (Rossi) LUOLearning Data Science and Big Data AnalyticsAnonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.comBlogger8125tag:blogger.com,1999:blog-5990946278123924979.post-27775061062572331422015-03-27T21:11:00.001-07:002015-03-27T21:11:29.329-07:00ABCD Initiative & LogoMy group got a new logo (see below) for the <a href="https://sites.google.com/site/xirossiluo/abcd" target="_blank">ABCD</a> (Huh? Yes!) initiative. There is a word hiding. Guess what "C" stands for?<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiX_uZcfTNjRSP51Cu8M_XGOd84em4XxEWiNno2GkTm1wNRrIw9Co6S9cSLvjMhk-BTEJHvvi9s-QArKEmhIbX6cb3eCR2DOJKiJYeh38gD5fIt6Y5MflF1FYYLA0U9jzaJtkUxAertDqQ/s1600/abcdLogo2.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiX_uZcfTNjRSP51Cu8M_XGOd84em4XxEWiNno2GkTm1wNRrIw9Co6S9cSLvjMhk-BTEJHvvi9s-QArKEmhIbX6cb3eCR2DOJKiJYeh38gD5fIt6Y5MflF1FYYLA0U9jzaJtkUxAertDqQ/s1600/abcdLogo2.png" height="124" width="320" /></a></div>
<br />Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-71329466928588802622015-03-16T11:13:00.003-07:002015-03-17T10:53:56.907-07:00Congratulations to Yi Zhao for winning an ENAR Distinguished Student Paper Award!Congratulations to Yi Zhao, a second year PhD student at <a href="https://www.facebook.com/BiostatisticsAtBrown" target="_blank">Brown Biostatistics</a>, who won an ENAR Distinguished Student Paper Award! This award is given by the <a href="http://www.enar.org/meetings/spring2015/" target="_blank">2015 ENAR meeting</a>. Yi presented our joint work on mediation analysis under unmeasured confounding (<a href="http://arxiv-web3.library.cornell.edu/abs/1410.7217" target="_blank">arXiv</a>). Hope you enjoyed her talk and all the wonderful talks this year!Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-78914539402719470512013-04-10T17:49:00.000-07:002013-04-10T17:51:57.419-07:00Postdoctoral Position AvailableWe have a postdoctoral position available on high dimensional analysis of phylogenetics data. This position is available through the <a href="http://www.brown.edu/research/projects/computational-molecular-biology/" target="_blank">Center for Computational and Molecular Biology</a> at Brown University in Providence, Rhode Island.
<br />
<br />
The successful candidate will collaborate with Casey Dunn (<a href="http://dunnlab.org/">http://dunnlab.org</a>) from the Department of Ecology and Evolutionary Biology, and Jean Wu (<a href="http://www.stat.brown.edu/zwu/">http://www.stat.brown.edu/zwu/</a>) and Rossi Luo (<a href="http://sites.google.com/site/xirossiluo/">http://sites.google.com/site/xirossiluo/</a>) from the Department of Biostatistics.
<br />
<br />
Research will focus on extending the ideas presented in the arXiv paper<br />
<a href="http://arxiv.org/abs/1302.2978"><span style="font-family: arial, sans-serif;">http://arxiv.org/abs/1302.2978</span></a><br />
<span style="font-family: inherit;"><br /></span>
<span style="font-family: inherit;">Funding is available for one year, with the possibility of additional funding for subsequent years. The position is available immediately.</span><br />
<span style="font-family: arial, sans-serif;"><br />
</span>Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-10248608325270380012012-04-26T19:01:00.000-07:002012-04-26T19:26:26.519-07:00What Causes Brain to Think?Brain causes us to behave as who we are, but what causes brain to think (activate)? Indirect measures of brain activities are usually preferred, such as fMRI, but the resulting statistics usually cannot resolve the causality issue. The simple box plots (image below, left) show no significant differences of the brain activity under two experimental conditions. Without modeling what the indirect response function looks like (e.g. <a href="http://en.wikipedia.org/wiki/Functional_magnetic_resonance_imaging#BOLD_hemodynamic_response" target="_blank">HRF</a>), our <a href="http://arxiv.org/abs/1110.6560" target="_blank">paper</a> (to appear in Journal of the American Statistical Association) provides a way to resolve the causality question: what causes brain to think?(image below, right)<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiItadjuy6oRxAdfAwN49WT6B3xVy6f3K7bJI9oUHGED4K9cWZaW3IIe8Cfo9cyZd7cnlqf2Dt_ALWjM-mY_yDHMyH2PBKul2tm7tZGdZzf-xruHVPE8DqZ3YamFSSvlu0Cgnb4eJxe7cI/s1600/Causal.001.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="480" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiItadjuy6oRxAdfAwN49WT6B3xVy6f3K7bJI9oUHGED4K9cWZaW3IIe8Cfo9cyZd7cnlqf2Dt_ALWjM-mY_yDHMyH2PBKul2tm7tZGdZzf-xruHVPE8DqZ3YamFSSvlu0Cgnb4eJxe7cI/s640/Causal.001.png" width="640" /></a></div>
<div class="separator" style="clear: both; text-align: left;">
This method is specifically designed to provide causal interpretation under interference, and is robust against any shape of the response function. The R package is <a href="http://cran.r-project.org/web/packages/cin/index.html" target="_blank">CIN</a>.</div>
<div>
<br /></div>Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-80582971912668829842012-04-23T19:31:00.000-07:002012-04-26T19:28:36.607-07:00New R Package SCIO PostedAs promised, the new R package for <a href="http://arxiv.org/abs/1203.3896" target="_blank">SCIO</a> (and <a href="http://rossiluo.blogspot.com/2012/03/statisticians-can-compute-fast.html" target="_blank">post</a>) has been uploaded to CRAN. It will be publicly available across all platforms in a few days.<br />
<br />
This is a version with the minimal set of functions for estimation. I will add more functionalities in future releases.<br />
<br />
I have also updated our R package <a href="http://cran.r-project.org/web/packages/clime/index.html" target="_blank">clime</a>. Thanks useRs for all the suggestions. The JASA paper is <a href="http://amstat.tandfonline.com/doi/abs/10.1198/jasa.2011.tm10155" target="_blank">here</a>, and the preprint is <a href="http://arxiv.org/abs/1102.2233" target="_blank">here</a>. I have also posted the cancer result using this method on this <a href="http://rossiluo.blogspot.com/2012/03/genetic-profiling-for-breast-cancer.html" target="_blank">post</a>.Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-27523020990719996232012-03-23T21:09:00.004-07:002012-04-26T19:28:57.738-07:00Statisticians Compute Fast<div class="separator" style="clear: both; text-align: left;">
Fast computation is increasingly important in analyzing data as we are living in an information explosion era. Here is another paper we submitted and the preprint is on <span class="Apple-style-span" style="font-family: Helvetica;"><span class="Apple-style-span" style="font-size: 15px;"><a href="http://arxiv.org/abs/1203.3896" target="_blank">arXiv 1203:3896</a>. An example in the paper is to compute the brain networks of ADHD children and controls. Here is a snap shot of the computation time.</span></span></div>
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiE0OgLzt3ArdeZaSEHUTB9sxcRUTKcHOWUd07lw1XmBhhjeBEMCBDEY8dVnO7GfFkHqRcxsedvzpRBcVy2WddrH-LmlEdavfu5bKUFQrM9TVSpJcEW7Kjd1Q-08WUsoYiwZzFVTzULpMQ/s1600/ADHDrunningtimeSnapshot.gif" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="620" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiE0OgLzt3ArdeZaSEHUTB9sxcRUTKcHOWUd07lw1XmBhhjeBEMCBDEY8dVnO7GfFkHqRcxsedvzpRBcVy2WddrH-LmlEdavfu5bKUFQrM9TVSpJcEW7Kjd1Q-08WUsoYiwZzFVTzULpMQ/s640/ADHDrunningtimeSnapshot.gif" width="640" /></a></div>
<div class="separator" style="clear: both; text-align: left;">
The new method is called SCIO. Will post the R package on CRAN.</div>Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-83684374138652297822012-03-20T20:17:00.007-07:002012-04-26T19:29:15.391-07:00Genetic Profiling for Breast Cancer TreatmentOur paper, <a href="http://pubs.amstat.org/doi/abs/10.1198/jasa.2011.tm10155" target="_blank">CLIME</a>, appeared in Journal of the American Statistical Association last year. It is always amazing to me that how science can help shape the future. Part of the paper employed a mathematical model to study if we can predict the outcome of a breast cancer patient based on her genetic profile. The prediction performance is interestingly good.<br />
<table align="center" cellpadding="0" cellspacing="0" class="tr-caption-container" style="margin-left: auto; margin-right: auto; text-align: center;"><tbody>
<tr><td style="text-align: center;"><a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjNqB_1JcbTGHp90VbkUy3vvRtP371BB5bwhghmPzkWw7Lem0Vrwhom9CHgTdIQ8EkDazzz2o-Adml_rGur29FBchHD42qlDmyihdcjLfHaNxxV3hdVNmzMQVKtAp9troHkZRgp7rnWDVQ/s1600/ClimeBreastCancerTable.gif" style="margin-left: auto; margin-right: auto;"><img border="0" height="320" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjNqB_1JcbTGHp90VbkUy3vvRtP371BB5bwhghmPzkWw7Lem0Vrwhom9CHgTdIQ8EkDazzz2o-Adml_rGur29FBchHD42qlDmyihdcjLfHaNxxV3hdVNmzMQVKtAp9troHkZRgp7rnWDVQ/s640/ClimeBreastCancerTable.gif" width="640" /></a></td></tr>
<tr><td class="tr-caption" style="text-align: center;">Average (SD) performance of classifying breast cancer treatment outcomes over 100 out-of-sample runs.</td></tr>
</tbody></table>
<div class="separator" style="clear: both; text-align: center;">
</div>
<div class="separator" style="clear: both; text-align: left;">
Both the sensitivity and specificity are high (>=75%). The R package for the CLIME method is <a href="http://cran.r-project.org/web/packages/clime/index.html" target="_blank">here</a> on CRAN, and I will provide an update very soon.</div>Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0tag:blogger.com,1999:blog-5990946278123924979.post-12412085173075986192011-11-20T07:36:00.015-08:002012-04-26T19:31:47.409-07:00New tools for investment portfolios?I posted a paper (<a href="http://arxiv.org/abs/1111.1133">arXiv: 1111.1133</a>) quite a long time ago on "High Dimensional Low Rank and Sparse Covariance Matrix Estimation via Convex Minimization". The software for this methodology, LOREC, is <a href="http://cran.r-project.org/web/packages/lorec/index.html" target="_blank">here</a> on CRAN.<br />
<br />
I still remember the stock market crash in 2009. This method is used to construct minimal risk stock portfolios. Here are the variances of monthly returns with different investment goals.<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEi37FdRciAaR9DtgJv63UbbOl5QouC0b0Ia1o8RSF4i5btFxUlvGu4j-jxbLbLjsII8_zaLJGg434y2bhaijrImiOuCTocI5VmHPxZ_sNP2xw2feiA5WFUlXhmC7yFJJr1TarUCvhT7AhM/s1600/SP100ReturnTable.gif" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="209" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEi37FdRciAaR9DtgJv63UbbOl5QouC0b0Ia1o8RSF4i5btFxUlvGu4j-jxbLbLjsII8_zaLJGg434y2bhaijrImiOuCTocI5VmHPxZ_sNP2xw2feiA5WFUlXhmC7yFJJr1TarUCvhT7AhM/s640/SP100ReturnTable.gif" width="640" /></a></div>
<div class="separator" style="clear: both; text-align: center;">
</div>
This new method, LOREC, performs favorably comparing with existing methods. Wish I had found this earlier in 2009.Anonymoushttp://www.blogger.com/profile/04852375240606278672noreply@blogger.com0