Mathematica 6.0: good choice statistics for an MD thesis?

*To*: mathgroup at smc.vnet.net*Subject*: [mg93179] Mathematica 6.0: good choice statistics for an MD thesis?*From*: Marc Heusser <marc.heusser at byeheusser.commercialspammers.invalid>*Date*: Thu, 30 Oct 2008 02:01:51 -0500 (EST)*Organization*: University of Zurich

I need to collect data, do statistical analyses and create graphics for the results for my MD thesis (suicides in elderly people). This means maybe 1500 cases with many variables known like age, residence address, religion, method etc (most nominal, not continuous variables). I would like to find clusters of similar cases, and find predictive factors etc. I have data on some 20 years. The data must be collected, ie exists in paper form only. I just got Mathematica 6, and found nice user interface options. Would you recommend using Mathematica to build a data collection mask (ie one case, with drop boxes for most data, checking eg birth date for validity etc), and collecting those 1500 cases with it in a notebook or external file - what data structure? Then I would need to combine this with some statistical base facts like number of inhabitants in a city (from census data). And then using the data for statistical tests to see whether we can prove an patterns, create clusters, and do a factor analysis of contributing factors? And last present the data in a graphically strong form to present results? If anyone has done parts of this task, especially created a small form to collect data in a structure that can be used for the statistical tests, I'd be very grateful for some example code to get me started. The same goes for data structures and running some built-in statistical tests. I have used Mathematica before, older versions, but for very different tasks, I have never done statistics in Mathematica before. Thank you in advance. Marc -- remove bye and from mercial to get valid e-mail <http://www.heusser.com>

**Follow-Ups**:**Re: Mathematica 6.0: good choice statistics for an MD thesis?***From:*"Mauricio Esteban Cuak" <cuak2000@gmail.com>