In <g class="gr_ gr_17 gr-alert gr_gramm gr_inline_cards gr_run_anim Punctuation only-ins replaceWithoutSep" id="17" data-gr-id="17">epidemiology</g> there are often many questions at issue. Any syllabus should contain methods for dealing with multiple questions. I would include p-value plots and adjusted p-values.
I would certainly include some of the effective methods for analysis of large <g class="gr_ gr_168 gr-alert gr_spell gr_inline_cards gr_run_anim ContextualSpelling ins-del" id="168" data-gr-id="168">data sets</g>, large N and small p. Recursive partitioning comes to mind.
For small N, large p, an alternative to PCA, non-negative matrix factorization, should be considered.
Cross-validation should be covered.
Statistical practice is coming under scrutiny. Consider two recent books.
Harris: Rigor Mortis.
Chandler: The seven deadly sins of psychology.
------------------------------
Sidney Young
Retired
------------------------------
Original Message:
Sent: 01-12-2018 12:42
From: Erick Suarez Perez
Subject: Computational statistics syllabus
Hi,
At the School of Public Health, University of Puerto Rico, we are preparing a proposal for a PhD program combined in Biostatistics ad Epidemiology. I wonder if someone could share a syllabus of Computational Statistics applied to Epidemiology at this level or send the link where i could find the aims, topics and references.
Regards,
Erick
Erick Suárez, PhD
Department of Biostatistics and Epidemiology
School of Public Health
University of Puerto Rico
------------------------------
Erick Suarez Perez
------------------------------