Ji Meng Loh -

TiddlyWiki\n[[Using this site]]\n

This is a summary of my work in this project. For more details, you can go to http://www.stat.columbia.edu/~meng/Astro\n\nAstronomers are interested in understanding the large scale structure that is evident in the universe. Various astrophysical theories have been developed to explain how such large scales of clustering could have developed.\n\nIn this project, we study the clustering properties of what are called absorption systems (or simply absorbers). These are believed to be non-luminous gas/dust clouds near very distant galaxies. They can be observed when they lie between the Earth and very bright quasi-stellar objects (~QSOs). By examining the spectra of light observed, the presence of these absorption systems, and their locations on the lines-of-sight to the ~QSOs can be determined. Understanding the clustering of absorption systems can (i) aid in understanding the true nature of absorption systems; (ii) serve as complements to studies of clustering of luminous matter.\n\nAn absorber catalog consists of the positions of absorbers on lines-of-sight. We developed a method to measure the degree of clustering of absorbers at various scales that includes information of absorber pairs from different lines-of-sight. It involves figuring out the correction factor needed to account for edge effects. Doing this can substantially reduce standard errors of estimates compared to more commonly used methods that obtain measures using only information of absorber pairs on the same line-of-sight.\n\nThis work has been published in Astrophysical Journal and the Journal of the American Statistical Association.\n\nFurther work involve (i) applying this method to the upcoming Sloan Digital Sky Survey (SDSS) absorber data. This dataset is 1000 times larger than the dataset we originally used; (ii) extending this method to estimate third-order moment characteristics. Doing this will allow us to examine the behavior of triplets of points. Current datasets are too small for this to be applicable, but the SDSS data will be large enough for third-order moment characteristics to be estimated. This can then be compared with similar measures of galaxies and stars.\n\n

The method of bootstrap is now a well-established method for statistic inference of independent data, so much so that it is often applied to dependent data (time series and spatial data). Theoretical justification of the boostrap for dependent data is often complex and involves assuming some kind of mixing condition that limits the range of dependence. A lot of this work was done by Kunsch, Lahari and others.\n\nMy work in this area is twofold: (1) developing a new resampling scheme and (2) understanding bootstrap of Gaussian random fields under fixed-domain asymptotics.\n\n1. We (Michael Stein and I) developed a modified resampling scheme, which we call the marked point method. Often, spatial bootstrap is done by block bootstrap, using windows to resample points. These windows/blocks are then joined together to form the new bootstrap sample. The use of windows is an attempt to capture the dependence present in the data (see Kunsch 1989). In the marked point method, we first assign to each point a mark. The sum of these marks give the statistic of interest. During the bootstrap procedure (and this can be done using blocks), the marks are resampled along with the points. By doing this the effect of joining independent blocks is reduced. \n\nThis work has been applied to an astronomy dataset, and has been published in Astrophysical Journal and Statistica Sincia.\n\n(Note: Kunsch in his 1989 paper, also introduced a block-of-blocks bootstrap that is very similar to the marked point bootstrap. It is, however, very rarely mentioned. This could be due to the complex notation. The formulation of the marked point bootstrap is very simple and makes it straightforward to implement. There is also additional flexibility with the marked point bootstrap, for example, in including corrections for edge effects.)\n\nFurther work include studying the behavior of this new method under various models of dependence; comparing with other bootstrap schemes, especially in the presence of edge effects; proving theoretical results.\n\n2. Most theoretical work done on the bootstrap of dependent data considers increasing domain asymptotics, i.e. the number of observations increase as the observation region increases. By assuming some sort of mixing condition together with an increasing domain, the blocks become almost independent. For random fields, it is not unusual to have a fixed observation region, and increasing number of observations in the region. An example is observations of the Cosmic Microwave Background (CMB). The (spherical) region is fixed, but as technology improves data at a finer and finer resolution becomes possible.\n\nWe cannot expect general theoretical results for the bootstrap in fixed domain asymptotics. This is because the dependence in the data does not decrease as the number of data point increase. We consider instead certain specific models for the correlation structure of a Gaussian random field in one dimension and show that the bootstrap can be consistent for some parameters when we difference the data to reduce the amount of dependence. We find that the amount of differencing needed depends on the smoothness of the random field. \n\nResults of this work will appear in Statistica Sinica.\n\nFurther work involve considering Gaussian random fields in two dimensions.

If you are thinking of statistics as a major or joint major, here are some links about careers in statistics that might be useful:\n** [[American Statistical Association|http://www.amstat.org/careers]] has information about careers in statistics. There are brochures that you can download. There is also a pdf file giving statistics about salaries.\n** [[Royal Statistical Society|http://www.rss.org.uk/main.asp?page=1999]] also has information about careers in statistcs. There is a powerpoint presentation that you can download. The RSS is based in the UK, so some information may not be relevant.\n** Do an internet search for e.g. "statistics career" and you will probably find many other useful sites. Some of these are maintained by professors in other universities.\n

Here are links to some of my collaborators' webpages:\n\n[[Regina Dolgoarshinnykh|http://www.stat.columbia.edu/~regina]]\n[[Yongtao Guan|http://info.med.yale.edu/eph/faculty/guan.html]]\n[[Woncheol Jang|http://jang.myweb.uga.edu/]]\n[[Naa Oyo Kwate|http://www.rci.rutgers.edu/~nokwate/]]\n[[Martin Lindquist|http://www.stat.columbia.edu/~martin]]\n[[Jingchen Liu|http://www.stat.columbia.edu/~jcliu]]\n[[Malgosia Madajewicz|http://www.sipa.columbia.edu/academics/directory/mm1174-fac.html]]\n[[Jean Quashnock|https://www.carthage.edu/directory/displayitem.cfm?id=336&fullview]]\n[[Michael Stein|http://www.stat.uchicago.edu/faculty/stein.html]]\n[[Ryan Yue|http://zicklin.baruch.cuny.edu/faculty/profiles/yu-ryan-yue]]\n[[Zhengyuan Zhu|http://www.unc.edu/~zhuz/zhu.html]]\n

[[Welcome]]\n[[My Contact]]\n

<<newTiddler>>\n<<saveChanges>>\n<<slider chkSliderOptionsPanel OptionsPanel 'options »' 'Change TiddlyWiki advanced options'>>\n

If you are a statistics major or joint-major, you are allowed to take some elective courses. These can be from any department as along as there is substantial application of statistical methodolgy that is not at the introductory level. If you are a joint major with say, math, advanced math content in the elective course would often be acceptable as well (both depts will need to agree to the elective). \n\nUnfortunately, there is no way I can keep track of all the courses that satisfy this criterion. You need to find out what the course teaches (from the syllabus, instructor etc), and then come see me for approval (bring along the syllabus and other relevant material e.g textbook borrowed from the library).

This course will provide an overview of the theory of spatial statistics and its applications, with more in-depth study of selected topics. We will also look at a few journal articles that deal with relatively recent developments in spatial statistical methodology. The focus will be on spatial point processes and (Gaussian) random fields. Specific areas that are covered include:\n\n''Spatial point processes''\nExploratory analysis of point processes – intensity, clustering, tests of complete randomness\nEstimating clustering – various measures; estimators of clustering; boundary effects\nMathematical theory for spatial point processes; mixing conditions\nPoint process models and model fitting\nDefining residuals after model fitting\nSimulation methods\nStatistical inference for inhomogeneous point processes\nBootstrap methods\nApplications – clustering in the universe ...\n\n''Gaussian Random Fields''\nVariogram estimation, variogram models\nKriging\nAsymptotic regimes in spatial statistics – increasing domain vs infill asymptotics; micro-ergodicity; infill asymptotics for estimating fractal dimension, in bootstrap; relating the two regimes\nModel-based geostatistics\nSimulation methods\nSampling design\nNon-stationary random fields\n\n''Other topics'' (time permitting)\nSpatial Scan Statistic – accounting for spatial correlation\nSpatial hazard modeling\nMarkov Random Fields\n\nThere will be one or two guest lectures introducing other fields in statistics where ideas from spatial statistics may be useful. One or two lecture slots may also be used for student presentations (either of journal articles, or of their research).\n\nNote: this course should be taken Pass/Fail

Type the text for 'Human and Social Dynamics'

This is not an exhaustive list, just what I found useful to know. I spent a bit of time summarizing this information and I am putting it here so I can find it again.\n\nArxiv.org - all IMS and Bernoulli Society(?) journals (some delay for current issue)\n\nIMS membership ($95, or $76 if renew early) gives full electronic access to all IMS journals including\nAnnals of applied statistics (AOAS)\nAnnals of Statistics\nStatistical Science\n\nJASA - free with ASA membership $125\nAmerican Statistician - free with ASA membership\nJournal of Agricultural and Environmental Statistics (JABES) $50\nJournal of Computation and Graphical Statistics (JCGS) $65\nTechnometrics $30\n\nJRSS B and C (JRSS) - one free with membership 82 pounds, 40 pounds each extra\nStatistica Sinica (ICSA) - free with ICSA membership $40\n\nScandinavian Journal of Statistics ($58 with ASA/IMS membership, else $68)\nBiometrika $68\nBiometrics - free with International Biometric Society membership $60\n\nToo expensive if not from institutional subscription!\nEnvironmetrics\nJournal of Statistical Planning and Inference (JSPI)\n\nBayesian Analysis - open access

/***\n|''Name:''|LegacyStrikeThroughPlugin|\n|''Description:''|Support for legacy (pre 2.1) strike through formatting|\n|''Version:''|1.0.1|\n|''Date:''|Jul 21, 2006|\n|''Source:''|http://www.tiddlywiki.com/#LegacyStrikeThroughPlugin|\n|''Author:''|MartinBudden (mjbudden (at) gmail (dot) com)|\n|''License:''|[[BSD open source license]]|\n|''CoreVersion:''|2.1.0|\n|''Browser:''|Firefox 1.0.4+; Firefox 1.5; InternetExplorer 6.0|\n\n***/\n\n//{{{\n\n// Ensure that the LegacyStrikeThrough Plugin is only installed once.\nif(!version.extensions.LegacyStrikeThroughPlugin)\n {\n version.extensions.LegacyStrikeThroughPlugin = true;\n\nconfig.formatters.push(\n{\n name: "legacyStrikeByChar",\n match: "==",\n termRegExp: /(==)/mg,\n element: "strike",\n handler: config.formatterHelpers.createElementAndWikify\n});\n\n} // end of "install only once"\n//}}}\n

\n[[Google calendar|http://www.google.com/calendar]]\n[[Google maps|http://maps.google.com]]\n[[Yahoo weather|http://weather.yahoo.com/forecast/USNY0996.html]]\n\n[[SentiWeb|http://www.sentiweb.org]]: Monitoring of disease incidence in France\n[[Stock data|http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html]] on Kenneth French's webpage\n\n[[Sloan Digital Sky Survey|http://www.sdss.org]]\n[[Center for AstroStatistics|http://astrostatistics.psu.edu/]] at Penn State

\n[[My Contact]]\n[[CV|jmcv.pdf]]\n[[Research]]\n{{indent{[[Papers]]\n{{indent{[[Talks]]\n{{indent{[[Astronomy]]\n\n[[Resources]]\n[[Links]]\n\n\n\n

[img[http://www.stat.columbia.edu/~meng/HomeFiles/jm2.jpg]]\nJi Meng Loh\n\nTel: 973-236-6230\nEmail: loh at research dot att dot com

<<search>><<closeAll>><<permaview>><<slider chkSliderEditing [[Editing Functions]] 'Editing functions »' 'Editing functions'>>\n\n\n\n\n\n

To make a tiddler that doesn't have a WikiWord as its name, you can enclose the name in double square brackets - edit this tiddler to see an example. After saving the tiddler you can then click on the link to create the new tiddler. NonWikiWordLinks permits tiddlers to be created with names that are made from character sets that don't have upper and lower case.

# [[Kwate, Loh, White and Saldana (2011)|http://www.stat.columbia.edu/~meng/Papers/jurbanhealth.Kwate.etal.pdf]] - Retail redlining in New York City: Racialized access to day-to-day retail resources - Journal of Urban Health (under revision).\n# Jang, Lim, Lazar, Loh, ~McDowell and Yu (2011) - Regression shrinkage and equality selection for highly correlated predictors with HORSES - submitted to Biometrics.\n# [[Yue, Lindquist and Loh (2011)|http://www.stat.columbia.edu/~meng/Papers/aoas.Yue.etal.2011.pdf]] - Meta-analysis of functional neuroimaging data using Bayesian nonparametric binary regression - accepted at Annals of Applied Statistics. \n# [[Yau and Loh (2011)|http://www.stat.columbia.edu/~meng/Papers/sinica.YauLoh.2011.pdf]] - A Generalization of the ~Neyman-Scott Process - accepted at Statistica Sinica.\n# [[Loh and Dasu (2011)|http://www.stat.columbia.edu/~meng/Papers/iciq.LohDasu.2011.pdf]] - Auditing Data Streams for Correlated Glitches - International Conference of Information Quality 2011.\n# [[Becker, Caceres, Hanson, Loh, Urbanek, Varshavsky and Volinsky (2011)|http://www.stat.columbia.edu/~meng/Papers/ubicomp.2011.pdf]] - Route classification using cellular handoff patterns - 13th ACM International Conference on Ubiquitous Computing (~UbiComp) 2011 (also at ~NetMob 2011).\n# [[Becker, Caceres, Hanson, Loh, Urbanek, Varshavsky and Volinsky (2011)|http://www.stat.columbia.edu/~meng/Papers/purba.2011.pdf]] - Clustering anonymized mobile call detail records to find usage groups - 1st Workshop on Pervasive Urban Applications (PURBA) 2011 (also at ~NetMob 2011).\n# [[Becker, Caceres, Hanson, Loh, Urbanek, Varshavsky and Volinsky (2011)|http://www.stat.columbia.edu/~meng/Papers/ieeepervasive.2011.pdf]] - A tale of one city: using cellular network data for urban planning - IEEE Pervasive Computing, 10, 18-26 (also at ~NetMob 2011).\n# [[Yue and Loh (2011)|http://www.stat.columbia.edu/~meng/Papers/biometrics.YueLoh.2011.pdf]] - Bayesian Nonparametric Intensity Estimation for Inhomogeneous Spatial Point processes - Biometrics, 67, 937-946.\n# [[Hariharan, Loh, Shanahan and Yamada (2010)|http://dl.acm.org/citation.cfm?id=1869863]] - Spatial Probabilistic Modeling of Calls to Businesses - ACM SIGSPATIAL 2010.\n# [[Kwate and Loh (2010)|http://www.sciencedirect.com/science/article/pii/S0091743510001751]] - Separate and Unequal: The Influence of Neighborhood and School Characteristics on Spatial Proximity between Fast Foods and Schools - Preventive Medicine, 51, 153-156.\n# [[Loh and Jang (2010)|http://www.stat.columbia.edu/~meng/Papers/JRSSC.LohJang.2010.pdf]] - Estimating a Cosmological Mass Bias Parameter with ~Semi-Parametric Bootstrap Bandwidth Selection - Journal of the Royal Statistical Society, Series C, 59, 761-779.\n# [[Loh (2010)|http://www.stat.columbia.edu/~meng/Papers/Kscan.12.submit.pdf]] - K-scan for Anomaly Detection in Disease Surveillance - Environmetrics, 22, 179-191.\n# [[Yue, Loh and Lindquist (2010)|http://www.stat.columbia.edu/~meng/Papers/SII.YueLohLindquist.2010.pdf]] - Adaptive Spatial Smoothing of fMRI images - Statistics and Its Interface, 3, 3-14.\n# [[Jang and Loh (2010)|http://www.stat.columbia.edu/~meng/Papers/AOAS.JangLoh.2010.pdf]] - Density Estimation for Grouped Data with Application to Line Transect Sampling - Annals of Applied Statistics, 4, 893-915.\n# [[Loh (2010)|http://www.stat.columbia.edu/~meng/Papers/Jspi.Loh.2010.pdf]] - Bootstrapping an Inhomogeneous Point Process - Journal of Statistical Planning and Inference, 140, 734-749.\n# [[Kwate, Yip, Loh and Williams (2009)|http://www.stat.columbia.edu/~meng/Papers/Kwate.08.pdf]] - Inequality in Obesigenic Environments: Fast Food Density in New York City - Health and Place, 15, 364-373\n# [[Lindquist, Loh, Atlas and Wager (2008)| http://dx.doi.org/10.1016/j.neuroimage.2008.10.065]] - Modeling the Hemodynamic Response Function in fMRI: Efficiency, Bias and Mis-modeling - ~NeuroImage, 45, ~S187-S198.\n# [[Loh, Lindquist and Wager (2008)|http://www.stat.columbia.edu/~meng/Papers/Sinica.Loh.08.pdf]] - Residual Analysis for Detecting Mismodeling in fMRI Images - Statistica Sinica, 18, 1421-1448.\n# [[Loh (2008)|http://www.stat.columbia.edu/~meng/Papers/ApJ.Loh08a.pdf]] - A Valid and Fast Spatial Bootstrap for Correlation Functions - Astrophysical Journal, 681, 726-734.\n#[[Loh (2008)|http://www.stat.columbia.edu/~meng/Papers/ApJ.Loh08.pdf]] - Estimating ~Third-Order Moments for an Absorber Catalog - Astrophysical Journal, 674, 636-643.\n# [[Loh and Stein (2008)|http://www3.stat.sinica.edu.tw/statistica/J18N2/J18N214/J18N214.html]] - Spatial Bootstrap with Increasing Observations in a Fixed Domain - Statistica Sinica, 18, 667-688\n# [[Guan and Loh (2007)|http://www.stat.columbia.edu/~meng/Papers/Jasa.Guan07.pdf]] - A thinned block Bootstrap Procedure for Modeling Inhomegeneous Spatial Point Patterns - Journal of the American Statistical Association, 102, 1377-1386.\n# [[Loh and Zhu (2007)|http://www.stat.columbia.edu/~meng/Papers/AOAS.Loh.07.pdf]] - Accounting for Spatial Correlation in the Scan Statistic - Annals of Applied Statistics, 1, 560-584.\n# [[Rzhetsky, Iossifov, Loh and White (2006)|http://www.stat.columbia.edu/~meng/Papers/PNAS.Rzhetsky06.pdf]] - Microparadigms: Chains of Collective Reasoning in Publications about Molecular Interactions - Proceedings of the National Academy of Science, 103, p4930-4945.\n# [[Loh and Stein (2004)|http://www.stat.columbia.edu/~meng/Papers/Sinica.Loh04.pdf]] - Bootstrapping a Spatial Point Process - Statistica Sinica, 14, p69-101.\n# [[Loh, Stein and Quashnock (2003)|http://www.stat.columbia.edu/~meng/Papers/Jasa.Loh03.pdf]] - Estimating the ~Large-Scale Structure of the Universe using ~Quasi-Stellar Object Carbon IV Absorbers - Journal of the American Statistical Association, 98, p522-532.\n# [[Loh, Quashnock and Stein (2001)|http://www.stat.columbia.edu/~meng/Papers/ApJ.Loh01.pdf]] - A Measurement of the ~Three-Dimensional Clustering of C IV ~Absorption-Line Systems on Scales of 5-1000 h^^-1^^ Mpc - Astrophysical Journal, 560, p606-612.\n# [[Stein, Quashnock and Loh (2000)|http://www.stat.columbia.edu/~meng/Papers/Annals.Stein00.pdf]] - Estimating the ~K-Function of a Point Process with an Application to Cosmology - Annals of Statistics, 28, p1508-1532.

My main area of interest is in spatial statistics. A main feature in spatial statistics is dependence or correlation between data points. The two main areas of spatial statistics are geostatistics and spatial point patterns.\n\nIn geostatistics, the data usually consists of observations within a region of a continuous random field, for example, measurements of ozone at monitoring stations. There is underlying variability in the observations at any one location, but more importantly, there is correlation between the observations at different locations. We can model the dependence structure and use the fitted model to make predictions at other locations. Sometimes the interest is in modeling the dependence of the observations on some measured covariates.\n\nIn spatial point patterns, the data consists of observations of objects in a region, e.g. positions of galaxies in space, of trees in a forest. Descriptive statistics would include the intensity of points and measure of clustering or inhibition of points. Various models, such as Gibbs model, log-Gaussian Cox processes, the ~Neymann-Scot model, can be used to model the observed point pattern. \n\nSome of my work is supported by NSF grants. Specifically, NSF #~AST-0507687 was awarded for work in the astronomy project. NSF #~SES-0624256 has just been awarded for the Human and Social Dynamics project.\n\n__Some current projects__:\n[[Astronomy]]\n[[Bootstrap of Spatial Data]]\n[[Scan statistic for correlated data]]\n[[fMRI]]\n[[Human and Social Dynamics]]\n\n[[Papers]]\n[[Collaborators]]

\n[[Stats java applets|http://www.stat.duke.edu/sites/java.html]] at Duke\n[[Data and Story Library (DASL)|http://lib.stat.cmu.edu/DASL/]] at CMU\n[[Statistical Computing Resources|http://www.ats.ucla.edu/stat/]] at UCLA\n\n[[JSTOR|http://www.jstor.org]]\n[[Arxiv.org|http://www.arxiv.org]]\n[[Project Euclid|http://projecteuclid.org]]\n\nInformation about statistics [[Journal]] subscription rates, which ones you get with memberships etc (correct as of Oct 2009)\n\n[[The R statistical package]]\n\n[[Numerical Recipes|http://www.nr.com]]\n\nGiving a talk? Robert Geroch has some [[suggestions|http://astro.uchicago.edu/home/web/lucia/research/geroch-talk]]. I have also collected a list of [[pointers]].\n

You can save a copy of this TiddlyWiki and edit it as you wish. Generic instructions are given below. There may be complications with some browsers. Check [[www.tiddlywiki.com|http://www.tiddlywiki.com]] for details.\n\n# right click on [[this link|empty.html]] and select 'Save link as...' or 'Save target as...'\n** do ''not'' try to use the File/Save command in your browser.\n** choose where to save the file, and what to call it (but keep the .HTML extension)\n# open the newly downloaded file in your browser\n# click the 'options' button on the right to set your username\n# edit, create and delete the tiddlers you want\n** you can change special tiddlers such as SiteTitle and MainMenu etc.\n# click the 'save changes' button on the right to save your changes\n# TiddlyWiki will make a backup copy of the existing file, and then replace it with the new version\n

The scan statistic was studied by Naus (1965) as a method to identify clusters in a one-dimensional point process. Kulldorff (1997) extended it to the spatial setting, so that clusters in a multidimensional point process can be identified. The Spatial Scan Statistic is widely used in epidemiology. The spatial scan statistic can be computed using the [[SatScan|http://www.satscan.org]] software.\n\nAn underlying assumption of the method is the independence between points. In certain cases, such as the occurrence of an infectious disease, some positive correlation between the occurrence of points is expected. Furthermore, in the Poisson model studied by Kulldorff (1997), there could be overdispersion during to unmeasured covariates or measurement error.\n\nIn a simulation study, we find that when there is positive correlation or overdispersion, there is an increased chance in finding significant clusters, resulting in more false alarms.\n\nWe developed a method to reduce the number of false alarms. This involves modeling any dependence found in the data using a spatial generalized linear model (Diggle and Tawn 1998). The fitted model is then used in the Monte Carlo step employed to obtain the p value of an identified cluster.\n

<<search>><<closeAll>><<permaview>><<newTiddler>><<newJournal 'DD MMM YYYY'>><<saveChanges>><<slider chkSliderOptionsPanel OptionsPanel 'options »' 'Change TiddlyWiki advanced options'>>

<<tabs txtMainTab Tags 'All tags' TabTags Timeline Timeline TabTimeline All 'All tiddlers' TabAll More 'More lists' TabMore>>\n

Ji Meng Loh

/***\nhttp://tiddlystyles.com/#theme:DevFire\n***/\n\n/*{{{*/\n#mainMenu{\n position: static;\n text-align: left;\n}\n#displayArea{\n margin: 0 1em 0 17em\n}\n\nbody {\nbackground: #000;\n}\n/*}}}*/\n/***\n!Link styles /% ============================================================= %/\n***/\n/*{{{*/\na,\na.button,\n#mainMenu a.button,\n#sidebarOptions .sliderPanel a{\n color: #ffbf00;\n border: 0;\n background: transparent;\n}\n\na:hover,\na.button:hover,\n#mainMenu a.button:hover,\n#sidebarOptions .sliderPanel a:hover\n#sidebarOptions .sliderPanel a:active{\n color: #ff7f00;\n border: 0;\n border-bottom: #ff7f00 1px dashed;\n background: transparent;\n text-decoration: none;\n}\n\n#displayArea .button.highlight{\n color: #ffbf00;\n background: #4c4c4c;\n}\n/*}}}*/\n/***\n!Header styles /% ============================================================= %/\n***/\n/*{{{*/\n.header{\n border-bottom: 2px solid #ffbf00;\n color: #fff;\n}\n\n.headerForeground a {\n color: #fff;\n}\n\n.header a:hover {\n border-bottom: 1px dashed #fff;\n}\n/*}}}*/\n/***\n!Main menu styles /% ============================================================= %/\n***/\n/*{{{*/\n#mainMenu {color: #fff;}\n#mainMenu h1{\n font-size: 1.1em;\n}\n#mainMenu li,#mainMenu ul{\n list-style: none;\n margin: 0;\n padding: 0;\n}\n/*}}}*/\n/***\n!Sidebar styles /% ============================================================= %/\n***/\n/*{{{*/\n#sidebar {\n left: 0;\n color: #fff;\n border: 2px solid #ffbf00;\n border-width: 0 2px 2px 0;\n}\n#sidebarOptions {\n background-color: #000;\n padding: 0;\n}\n\n#sidebarOptions a{\n margin: 0;\n color: #ffbf00;\n border: 0;\n}\n#sidebarOptions a:hover {\n color: #4c4c4c;\n background-color: #ffbf00;\n\n}\n\n#sidebarOptions a:active {\n color: #ffbf00;\n background-color: transparent;\n}\n\n#sidebarOptions .sliderPanel {\n background-color: #333;\n margin: 0;\n}\n\n#sidebarTabs {background-color: #4c4c4c;}\n#sidebarTabs .tabSelected {\n padding: 3px 3px;\n cursor: default;\n color: #ffbf00;\n background-color: #666;\n}\n#sidebarTabs .tabUnselected {\n color: #ffbf00;\n background-color: #5f5f5f;\n padding: 0 4px;\n}\n\n#sidebarTabs .tabUnselected:hover,\n#sidebarTabs .tabContents {\n background-color: #666;\n}\n\n.listTitle{color: #FFF;}\n#sidebarTabs .tabContents a{\n color: #ffbf00;\n}\n\n#sidebarTabs .tabContents a:hover{\n color: #ff7f00;\n background: transparent;\n}\n\n#sidebarTabs .txtMoreTab .tabSelected,\n#sidebarTabs .txtMoreTab .tab:hover,\n#sidebarTabs .txtMoreTab .tabContents{\n color: #ffbf00;\n background: #4c4c4c;\n}\n\n#sidebarTabs .txtMoreTab .tabUnselected {\n color: #ffbf00;\n background: #5f5f5f;\n}\n\n.tab.tabSelected, .tab.tabSelected:hover{color: #ffbf00; border: 0; background-color: #4c4c4c;cursor:default;}\n.tab.tabUnselected {background-color: #666;}\n.tab.tabUnselected:hover{color:#ffbf00; border: 0;background-color: #4c4c4c;}\n.tabContents {\n background-color: #4c4c4c;\n border: 0;\n}\n.tabContents .tabContents{background: #666;}\n.tabContents .tabSelected{background: #666;}\n.tabContents .tabUnselected{background: #5f5f5f;}\n.tabContents .tab:hover{background: #666;}\n/*}}}*/\n/***\n!Message area styles /% ============================================================= %/\n***/\n/*{{{*/\n#messageArea {background-color: #666; color: #fff; border: 2px solid #ffbf00;}\n#messageArea a:link, #messageArea a:visited {color: #ffbf00; text-decoration:none;}\n#messageArea a:hover {color: #ff7f00;}\n#messageArea a:active {color: #ff7f00;}\n#messageArea .messageToolbar a{\n border: 1px solid #ffbf00;\n background: #4c4c4c;\n}\n/*}}}*/\n/***\n!Popup styles /% ============================================================= %/\n***/\n/*{{{*/\n.popup {color: #fff; background-color: #4c4c4c; border: 1px solid #ffbf00;}\n.popup li.disabled{color: #fff;}\n.popup a {color: #ffbf00; }\n.popup a:hover { background: transparent; color: #ff7f00; border: 0;}\n.popup hr {color: #ffbf00; background: #ffbf00;}\n/*}}}*/\n/***\n!Tiddler Display styles /% ============================================================= %/\n***/\n/*{{{*/\n.title{color: #fff;}\nh1,h2,h3,h4,h5 {\n color: #fff;\n background-color: transparent;\n border-bottom: 1px solid #333;\n}\n\n.subtitle{\n color: #666;\n}\n\n.viewer {color: #fff; }\n\n.viewer table{background: #666; color: #fff;}\n\n.viewer th {background-color: #996; color: #fff;}\n\n.viewer pre, .viewer code {color: #ddd; background-color: #4c4c4c; border: 1px solid #ffbf00;}\n\n.viewer hr {color: #666;}\n\n.tiddler .button {color: #4c4c4c;}\n.tiddler .button:hover { color: #ffbf00; background-color: #4c4c4c;}\n.tiddler .button:active {color: #ffbf00; background-color: #4c4c4c;}\n\n.toolbar {\n color: #4c4c4c;\n}\n\n.toolbar a.button,\n.toolbar a.button:hover,\n.toolbar a.button:active,\n.editorFooter a{\n border: 0;\n}\n\n.footer {\n color: #ddd;\n}\n\n.selected .footer {\n color: #888;\n}\n\n.highlight, .marked {\n color: #000;\n background-color: #ffe72f;\n}\n.editorFooter {\n color: #aaa;\n}\n\n.tab{\n-moz-border-radius-topleft: 3px;\n-moz-border-radius-topright: 3px;\n}\n\n.tagging,\n.tagged{\n background: #4c4c4c;\n border: 1px solid #4c4c4c; \n}\n\n.selected .tagging,\n.selected .tagged{\n background-color: #333;\n border: 1px solid #ffbf00;\n}\n\n.tagging .listTitle,\n.tagged .listTitle{\n color: #fff;\n}\n\n.tagging .button,\n.tagged .button{\n color: #ffbf00;\n border: 0;\n padding: 0;\n}\n\n.tagging .button:hover,\n.tagged .button:hover{\nbackground: transparent;\n}\n\n.selected .isTag .tagging.simple,\n.selected .tagged.simple,\n.isTag .tagging.simple,\n.tagged.simple {\n float: none;\n display: inline;\n border: 0;\n background: transparent;\n color: #fff;\n margin: 0;\n}\n\n.cascade {\n background: #4c4c4c;\n color: #ddd;\n border: 1px solid #ffbf00;\n}\n/*}}}*/

/*{{{*/\n* html .tiddler {\n height: 1%;\n}\n\nbody {\n font-size: .75em;\n font-family: arial,helvetica;\n margin: 0;\n padding: 0;\n}\n\nh1,h2,h3,h4,h5 {\n font-weight: bold;\n text-decoration: none;\n padding-left: 0.4em;\n}\n\nh1 {font-size: 1.35em;}\nh2 {font-size: 1.25em;}\nh3 {font-size: 1.1em;}\nh4 {font-size: 1em;}\nh5 {font-size: .9em;}\n\nhr {\n height: 1px;\n}\n\na{\n text-decoration: none;\n}\n\ndt {font-weight: bold;}\n\nol { list-style-type: decimal }\nol ol { list-style-type: lower-alpha }\nol ol ol { list-style-type: lower-roman }\nol ol ol ol { list-style-type: decimal }\nol ol ol ol ol { list-style-type: lower-alpha }\nol ol ol ol ol ol { list-style-type: lower-roman }\nol ol ol ol ol ol ol { list-style-type: decimal }\n\n.txtOptionInput {\n width: 11em;\n}\n\n#contentWrapper .chkOptionInput {\n border: 0;\n}\n\n.externalLink {\n #text-decoration: underline;\n}\n\n.indent {margin-left:3em;}\n.outdent {margin-left:3em; text-indent:-3em;}\ncode.escaped {white-space:nowrap;}\n\n.tiddlyLinkExisting {\n font-weight: bold;\n}\n\n.tiddlyLinkNonExisting {\n font-style: italic;\n}\n\n/* the 'a' is required for IE, otherwise it renders the whole tiddler a bold */\na.tiddlyLinkNonExisting.shadow {\n font-weight: bold;\n}\n\n#mainMenu .tiddlyLinkExisting, \n#mainMenu .tiddlyLinkNonExisting,\n#sidebarTabs .tiddlyLinkNonExisting{\n font-weight: normal;\n font-style: normal;\n}\n\n#sidebarTabs .tiddlyLinkExisting {\n font-weight: bold;\n font-style: normal;\n}\n\n.header {\n position: relative;\n}\n\n.header a:hover {\n background: transparent;\n}\n\n.headerShadow {\n position: relative;\n padding: 2em 0em 1em 1em;\n left: -1px;\n top: -1px;\n}\n\n.headerForeground {\n position: absolute;\n padding: 2em 0em 1em 1em;\n left: 0px;\n top: 0px;\n}\n\n.siteTitle {\n font-size: 2em;\n}\n\n.siteSubtitle {\n font-size: 1.2em;\n}\n\n#mainMenu {\n position: absolute;\n left: 0;\n width: 10em;\n text-align: right;\n line-height: 1.6em;\n padding: 1.5em 0.5em 0.5em 0.5em;\n font-size: 1.1em;\n}\n\n#sidebar {\n position: absolute;\n right: 3px;\n width: 16em;\n font-size: .9em;\n}\n\n#sidebarOptions {\n padding-top: 0.3em;\n}\n\n#sidebarOptions a {\n margin: 0em 0.2em;\n padding: 0.2em 0.3em;\n display: block;\n}\n\n#sidebarOptions input {\n margin: 0.4em 0.5em;\n}\n\n#sidebarOptions .sliderPanel {\n margin-left: 1em;\n padding: 0.5em;\n font-size: .85em;\n}\n\n#sidebarOptions .sliderPanel a {\n font-weight: bold;\n display: inline;\n padding: 0;\n}\n\n#sidebarOptions .sliderPanel input {\n margin: 0 0 .3em 0;\n}\n\n#sidebarTabs .tabContents {\n width: 15em;\n overflow: hidden;\n}\n\n.wizard {\n padding: 0.1em 0em 0em 2em;\n}\n\n.wizard h1 {\n font-size: 2em;\n font-weight: bold;\n background: none;\n padding: 0em 0em 0em 0em;\n margin: 0.4em 0em 0.2em 0em;\n}\n\n.wizard h2 {\n font-size: 1.2em;\n font-weight: bold;\n background: none;\n padding: 0em 0em 0em 0em;\n margin: 0.2em 0em 0.2em 0em;\n}\n\n.wizardStep {\n padding: 1em 1em 1em 1em;\n}\n\n.wizard .button {\n margin: 0.5em 0em 0em 0em;\n font-size: 1.2em;\n}\n\n#messageArea {\nposition:absolute; top:0; right:0; margin: 0.5em; padding: 0.5em;\n}\n\n*[id='messageArea'] {\nposition:fixed !important; z-index:99;}\n\n.messageToolbar {\ndisplay: block;\ntext-align: right;\n}\n\n#messageArea a{\n text-decoration: underline;\n}\n\n.popup {\n font-size: .9em;\n padding: 0.2em;\n list-style: none;\n margin: 0;\n}\n\n.popup hr {\n display: block;\n height: 1px;\n width: auto;\n padding: 0;\n margin: 0.2em 0em;\n}\n\n.listBreak {\n font-size: 1px;\n line-height: 1px;\n}\n\n.listBreak div {\n margin: 2px 0;\n}\n\n.popup li.disabled {\n padding: 0.2em;\n}\n\n.popup li a{\n display: block;\n padding: 0.2em;\n}\n\n.tabset {\n padding: 1em 0em 0em 0.5em;\n}\n\n.tab {\n margin: 0em 0em 0em 0.25em;\n padding: 2px;\n}\n\n.tabContents {\n padding: 0.5em;\n}\n\n.tabContents ul, .tabContents ol {\n margin: 0;\n padding: 0;\n}\n\n.txtMainTab .tabContents li {\n list-style: none;\n}\n\n.tabContents li.listLink {\n margin-left: .75em;\n}\n\n#displayArea {\n margin: 1em 17em 0em 14em;\n}\n\n\n.toolbar {\n text-align: right;\n font-size: .9em;\n visibility: hidden;\n}\n\n.selected .toolbar {\n visibility: visible;\n}\n\n.tiddler {\n padding: 1em 1em 0em 1em;\n}\n\n.missing .viewer,.missing .title {\n font-style: italic;\n}\n\n.title {\n font-size: 1.6em;\n font-weight: bold;\n}\n\n.missing .subtitle {\n display: none;\n}\n\n.subtitle {\n font-size: 1.1em;\n}\n\n.tiddler .button {\n padding: 0.2em 0.4em;\n}\n\n.tagging {\nmargin: 0.5em 0.5em 0.5em 0;\nfloat: left;\ndisplay: none;\n}\n\n.isTag .tagging {\n#display: block;\ndisplay: none;\n}\n\n.tagged {\nmargin: 0.5em;\nfloat: right;\ndisplay:none;\n}\n\n.tagging, .tagged {\nfont-size: 0.9em;\npadding: 0.25em;\n}\n\n.tagging ul, .tagged ul {\nlist-style: none;margin: 0.25em;\npadding: 0;\n}\n\n.tagClear {\nclear: both;\n}\n\n.footer {\n font-size: .9em;\n}\n\n.footer li {\ndisplay: inline;\n}\n\n* html .viewer pre {\n width: 99%;\n padding: 0 0 1em 0;\n}\n\n.viewer {\n line-height: 1.4em;\n padding-top: 0.5em;\n}\n\n.viewer .button {\n margin: 0em 0.25em;\n padding: 0em 0.25em;\n}\n\n.viewer blockquote {\n line-height: 1.5em;\n padding-left: 0.8em;\n margin-left: 2.5em;\n}\n\n.viewer ul, .viewer ol{\n margin-left: 0.5em;\n padding-left: 1.5em;\n}\n\n.viewer table {\n border-collapse: collapse;\n margin: 0.8em 1.0em;\n}\n\n.viewer th, .viewer td, .viewer tr,.viewer caption{\n padding: 3px;\n}\n\n.viewer table.listView {\n font-size: 0.85em;\n margin: 0.8em 1.0em;\n}\n\n.viewer table.listView th, .viewer table.listView td, .viewer table.listView tr {\n padding: 0px 3px 0px 3px;\n}\n\n.viewer pre {\n padding: 0.5em;\n margin-left: 0.5em;\n font-size: 1.2em;\n line-height: 1.4em;\n overflow: auto;\n}\n\n.viewer code {\n font-size: 1.2em;\n line-height: 1.4em;\n}\n\n.editor {\nfont-size: 1.1em;\n}\n\n.editor input, .editor textarea {\n display: block;\n width: 100%;\n font: inherit;\n}\n\n.editorFooter {\n padding: 0.25em 0em;\n font-size: .9em;\n}\n\n.editorFooter .button {\npadding-top: 0px; padding-bottom: 0px;}\n\n.fieldsetFix {border: 0;\npadding: 0;\nmargin: 1px 0px 1px 0px;\n}\n\n.sparkline {\n line-height: 1em;\n}\n\n.sparktick {\n outline: 0;\n}\n\n.zoomer {\n font-size: 1.1em;\n position: absolute;\n padding: 1em;\n}\n\n.cascade {\n font-size: 1.1em;\n position: absolute;\n overflow: hidden;\n}\n/*}}}*/

\nJSM 2010\nIWAP 2010\n\nJSM 2009\n\nFACM 2008\n\nJSM 2007: [[Clustering of Absorptions Systems|http://www.stat.columbia.edu/~meng/Talks/jsm2007_loh.pdf]]

Fall 2008\n[[W4105: Probability]]\n\nSpring 2008\n[[W1211: Introduction to Statistics (B)]] - Sections 1 and 3 only.\n[[G8220: Topics in Spatial Statistics]] - Wednesdays 2:10-4:00 pm\n\n[[The R statistical package]]

Official R [[website|http://www.r-project.org]]\n[[Link|http://cran.r-project.org/manuals.html]] to R manuals\nR [[reference card|http://www.stat.columbia.edu/~meng/Reference/Rrefcard.pdf]] containing 4 pages of useful commands.\n[[Statsci.org|http://www.statsci.org/r.html]] webpage on R\n[[Webpage|http://www.stats.ox.ac.uk/pub/MASS3/]] for the 3rd edition of the Modern Applied Statistics with S book by Venables and Ripley. The online [[complements|http://www.stats.ox.ac.uk/pub/MASS3/Compl.shtml]] has a section for using the book with R.\n\n__Some R tutorials on the web__:\n#[[Resources to help you learn and use R|http://www.ats.ucla.edu/stat/r/]]\n#[[Using R|http://mercury.bio.uaf.edu/mercury/R/R.html]]\n#[[R tutorial at Union College|http://www.cyclismo.org/tutorial/R]]\n#[[R tutorial at Illinois State U|http://www.math.ilstu.edu/dhkim/Rstuff/Rtutor.html]]\n#[[Statistics with R|http://zoonek2.free.fr/UNIX/48_R/all.html]]\n#[[R intro|http://www.stat.berkeley.edu/~spector/R.pdf]]\n#[[Econometrics in R|http://cran.r-project.org/doc/contrib/Farnsworth-EconometricsInR.pdf]]

A 'tiddler' is the name given to a unit of microcontent in TiddlyWiki.\n\nTiddlers are pervasive in TiddlyWiki. The MainMenu is defined by a tiddler, plugins are delivered in tiddlers, there are special StyleSheet tiddlers, and so on.\n\nOther systems have analogous concepts with more prosaic names: like "items", "entries", "entities". Even though "tiddler" is undoubtedly a silly name it at least has the virtue of being confusingly distinctive rather than confusingly generic.

TiddlyWiki is a free microcontent WikiWikiWeb created by Jeremy Ruston. It's written in HTML, CSS and ~JavaScript to run on any modern browser without needing any server side logic. It allows anyone to create personal self-contained hypertext documents that can be posted to a webserver, sent by email or kept on a USB thumb drive to make a Wiki on a stick. This is revision <<version>> of TiddlyWiki, and is published under an ~OpenSource License.\n\nA TiddlyWiki is like a blog because it's divided up into neat little chunks, but it encourages you to read it by hyperlinking rather than sequentially: if you like, a non-linear blog analogue that binds the individual microcontent items into a cohesive whole. There are also several TiddlyWiki adaptations by other developers based on earlier versions.\n\nOne of the neatest features of TiddlyWiki is that it is entirely self-contained in a single HTML file - even including graphics. The file contains the actual hypertext document, and the ~JavaScript, cascading stylesheets and HTML necessary to both view and edit it. This means that it is trivial to host a TiddlyWiki on a website, or to distribute one by email. And anyone with a reasonably recent web browser will be able to read and edit it.

If you have taken a probability/statistics course elsewhere and think you can transfer credit for a W1111 or W1211 requirement, get a form from the envelope on my office door (or download from [[here|http://www.stat.columbia.edu/~meng/HomeFiles/TransferCreditForm.pdf]]). \n\nFill up the form, put it in an envelope with whatever supporting documents you have (e.g. syllabus of the course taken, transcript, old hws/exams/project work, textbook), and put the envelope in my mailbox or under my door. The idea is to provide as much information as you can so that I can ascertain whether the material you learnt is similar to that taught in W1111/W1211. You don't need to bring everthing listed, but the more the better (just a course description is NOT enough). \n\nYou would sometimes need to take a statistics exam at the end of the semester with the current W1111/W1211 class.

# If you need to contact me or meet with me, please send me an email. Most questions can be answered by email. Otherwise we can arrange an appointment.\n# Very often students ask me questions about information that is actually readily available:\n** College bulletin: Copies are available from the [[Registrar|http://www.columbia.edu/cu/registrar]] and available online [[here|http://www.college.columbia.edu/bulletin/]]. Please refer to the relevant sections BEFORE coming to see me. I often refer to it myself to answer students' questions.\n** The [[Statistics|http://www.stat.columbia.edu]] dept website has information about statistics courses offered by the dept, textbooks and class meeting times. There is also a link to the biostatistics dept in the uptown campus.\n** These departments have joint major programs with statistics: [[Economics|http://www.columbia.edu/cu/economics]], [[Political science|http://www.columbia.edu/cu/polisci]], [[Mathematics|http://www.math.columbia.edu]].\n** The [[Columbia College advising page|http://www.college.columbia.edu/advising]] is a comprehensive site with information about all the departments. Very useful for help in choosing a major.\n** The [[Registrar|http://www.columbia.edu/cu/registrar/]] website has lots of information relevant to students, including the academic calendar, directory of classes and much more.\n[[Transfer of Credit]]\n[[Electives]]\n[[Career in Statistics]]\n\nSeveral courses offered by the statistics dept have very similar names and descriptions, because they cover roughly the same material, but with different depth. I have listed below the courses in increasing level of difficulty:\nW1001\nW1111\nW1211\nW4150\n(W3000 and W4107), or (W4105 and W4107) or W4109\n\nHere is a bit more information:\n* W1001: Elementary introduction to statistical concepts, with very little or no math.\n* W1111: Introduction to probability and statistics that does not require any calculus background (Material taught is about the same as W1211). Satisfies statistics requirement for some (BUT NOT ALL) majors - check the bulletin.\n* W1211: Introduction to probability and statistics that requires at least a semester of calculus. Material covered is about the same as W1111, but at a slightly higher level. Satisfies statistics requirement for most majors (check the bulletin).\n* W4150: Material taught is similar to W1111/W1211 but at a slightly higher level (e.g. more proofs). Taken by some non-statistics masters students (e.g. SIPA). If you need 4105/4107 for your major, you probably don't need to take this.\n* W3000 and W4105: Both are probability courses, and cover roughly the same material. W3000 is open only to undergraduates, but currently (as of Jan 2006) is only offered in the Fall semesters. W4105 is open to Masters students. If one course satisfies a major requirement, chances are the other course will too (but please check the bulletin - requirements are regularly modified).\n* W4105/W4107/W4109: these are advanced undergraduate and beginning masters level courses, and provide the theoretical foundation for the more advanced statistics courses. Probability theory is taught in W4105, and statistical inference in W4107. W4105 is a pre-requisite for W4107. Some instructors allow students to take both concurrently, but some don't. Check with the instructor involved. W4109 is a semester-long course that meets twice the usual amount of hours and is W4105 (first half of semester) and W4107 (second half of semester) combined. (Note: For the ~Econ-Math major, W4109 cannot be used in place of W4105 and W4107.)\n* Most, if not all, the statistics courses above the W4200 level can be taken in any order. These should usually be taken only after the 4105-4107 sequence.\nIf you have any doubts about what course to take, please check with your major advisor, or with me.

Within the main story column you can click on bold links to read a linked tiddler. When you hover the mouse over a tiddler several toolbar buttons appear. You can edit the text of any tiddler by double-clicking on it (or selecting 'edit' from the toolbar), but your changes won't get saved permanently until you make your own copy of TiddlyWiki, as described in SaveChanges.\n\nWhen you click on a tiddler link within another tiddler, the new one is opened immediately below the current one. If the target tiddler is already open, TiddlyWiki just uses smooth scrolling to bring it into view. More or less the same thing happens when clicking on a tiddler link within the menu or sidebar, except that the tiddler opens at the top of the page if it is not already open.

This course has a calculus pre-requisite. It is currently a requirement for the Economics major.\n\nLog on to [[Courseworks|https://courseworks.columbia.edu]] for course information, syllabus, handouts, hws and solutions.\n\n[[Lecture Summary|http://www.stat.columbia.edu/~meng/W1211/1211Summary.html]]\n\n[[The R statistical package]]

Hi, welcome to my homepage. Click on the links on the left to see the various sections of this webpage.\n\nNote: I am not longer with Columbia. If you have questions about the Statistics major, you should contact Professor Rabinowitz at dan@stat.columbia.edu

A Wiki is a popular way of building collaborative websites. It's based on the two ideas of allowing users to easily edit any page and the use of special WikiWord notation to automagically create links between pages. See Wikipedia for [[more details|http://en.wikipedia.org/wiki/Wiki]].\n\nTiddlyWiki is fundamentally different from a conventional Wiki because it is not based on separate, entire pages of content, but rather items of microcontent referred to as [[Tiddlers|Tiddler]] that live together on the same page.\n\nOut of the box, TiddlyWiki is also different because it doesn't support public editing - or indeed any persistent editing when viewed over the web. However, there are several TiddlyWiki adaptations and Plugins that provide these features in a wide range of different configurations.

A WikiWord is a word composed of a bunch of other words slammed together with each of their first letters capitalised. WikiWord notation in a conventional WikiWikiWeb is used to name individual pages while TiddlyWiki uses WikiWord titles for smaller chunks of microcontent. Referring to a page with a WikiWord automatically creates a link to it. Clicking on a link jumps to that page or, if it doesn't exist, to an editor to create it. It's also easy to have NonWikiWordLinks, and there's a WikiWordEscape for situations where you don't want a WikiWord to be interpreted as a link.

Sometimes it's handy to be able to write WikiWords without them being recognised as links (for people's names, for instance). You can do this by preceding the WikiWord with a tilde ({{{~}}}). For example, ~JamesBond, ~JavaScript and ~TiddlyWiki

In fMRI, the brain of a subject is scanned while he/she responds to some experimental stimuli. The brain is divided into voxels and for each voxel, there is a time series of measurements indicating the response of that part of the brain to the stimuli. A regression is typically performed for each voxel, with the regressor being some model representing the experimental stimuli.\n\nIn this project, Martin Lindquist and I are concerned with model checking and identifying model misfit. We do this by studying the residuals of the regression (as in typically done in other statistical applications). Specifically, we do this by using some sort of scan statistic to look for peaks in consecutive residuals. A Monte Carlo test can be done to check the significance of the scan statistic (and indicative of model misfit). We also looked at using Sidak's inequality as a non-computationally intensive method to obtain upper bounds on the p values.\n\nWe studied various types of model misfit and examined the power of such a test under these different types of misfit. We also worked out expressions for the bias of the regression estimates as well as how the regular regression t test is affected when the assumed predictor model is incorrect.\n\nOther related projects include extending this to multiple independent runs and to the spatial setting involving multiple voxels.

These pointers I collected from a variety of sources (articles, books etc). They are not rules, merely guidelines. They are not in any particular order. I find it helpful to go through these every now and then, because I tend to forget!\n\n# Make sure fonts and figures are not too small; do not put too many lines of text.\n# Don't show scans of your papers.\n# Don't use red or green - people who are color-blind can't see these colors.\n# Don't put too many equations on a page; your talk shouldn't focus too much on the technical details.\n# If you show a figure, describe it clearly and in detail.\n# Provide essential background and definitions.\n# Avoid unnecessarily fanciful transitions, clip art ...\n# Speak clearly, loud enough, and not too quickly.\n# Try not to cram too much into a single talk. A commonly expressed guideline is 1 slide for every 2 minutes (although I have heard excellent talks that break this "rule"). Have one (good) main idea and develop it.\n# The first few slides (first 5 minutes?) are important - this is where you gain or lose the interest of your audience. Start strong: be enthusiastic; describe an interesting motivating example ...\n# use sans serif fonts (serif fonts have connecting strokes that make characters flow together - they may be good for books, but make words difficult to read on the screen.)\n# Some people tend to let their voice drop off towards the end of a sentence. Makes it difficult to hear what they are saying.\n# Practice - vocalizing your ideas/thoughts helps to clarify your thinking and to keep your talk running smoothly. You also get an idea of the length of your talk.\n# Be organized - if you will be referring to a slide often, set it aside after use (for transparencies), or have copies of that slide at appropriate locations in your presentation so that you don't have to scroll through your Powerpoint presentation looking for that one slide..\n# Do a quick run-through of your Powerpoint presentation e.g. to make sure that sounds/movies work.\n# Keep copies of files in various places - e.g. in your laptop, in a couple of usb drives, on a server.\n# If you are using an overhead transparency projector, be aware of where you are standing and whether your arms or shoulders are blocking the light.\n# End on time - don't expect others to be as interested in your work as you are.\n