> Some scientists think that’s what’s really needed: for journalists to simply learn more about statistics in order to better weigh the validity of new studies as they report on them.
Monthly Archives: July 2015
Here is a careful recap of the journalistic coverage of Iraq mortality estimation
Comments Off on Here is a careful recap of the journalistic coverage of Iraq mortality estimation
Filed under Uncategorized
How do we stop data errors?
On my mind, due to some careful checking of results prior to submissions. Nothing substantive changed, but the numbers have to be right.
Correction
Data Errors in Results Section
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.6076
Data Errors in Table 2
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.6072
Data Errors in Table 3 and in Text
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.6074
Incorrect Data in Table 4
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.5464
Filed under Uncategorized
I missed SciPy 2015, what did I miss?
Hours of video available online: http://www.analyticsvidhya.com/blog/2015/07/data-science-videos-scipy-2015/
Comments Off on I missed SciPy 2015, what did I miss?
Filed under software engineering
ADVI in PyMC3
I don’t know how to use it and I don’t know if it works yet, but I’m watching the demos grow: https://github.com/pymc-devs/pymc3/blob/2250386c826e8744e13654188074dd69dad829de/pymc3/examples/advi.ipynb https://github.com/pymc-devs/pymc3/blob/01312efb319f45ec81824f7eccfea69e7b5a7ba1/pymc3/examples/GLM-hierarchical-ADVI.ipynb
I think I want some
Comments Off on ADVI in PyMC3
Filed under Uncategorized
Uncertainty quantification by clinicians
I continue to be drawn to articles in JAMA Oncology about how we wrestle with uncertain information in life-and-death decision making:
http://oncology.jamanetwork.com/article.aspx?articleID=2398178
http://oncology.jamanetwork.com/article.aspx?articleID=2398176
Comments Off on Uncertainty quantification by clinicians
Filed under Uncategorized
Laplace approximation in PyMC3, revisited
I found an even better example of the value of Laplace approximation, and its just a small tweak to the example I did a few weeks ago: http://nbviewer.ipython.org/gist/aflaxman/6d0a9ff2441348f3a130
Comments Off on Laplace approximation in PyMC3, revisited
Filed under statistics
Just when I get busy, a pile of interesting NAP reports
They’ve also made it extra annoying to download them, but I won’t let that stop me:
Cross-Sector Responses to Obesity: Models for Change: Workshop Summary http://www.nap.edu/catalog/21706/cross-sector-responses-to-obesity-models-for-change-workshop-summary
Preventing Intimate Partner Violence in Uganda, Kenya, and Tanzania: Summary of a Joint Workshop by the Institute of Medicine, the National Research Council, and the Ugandan National Academy of Sciences http://www.nap.edu/catalog/21756/preventing-intimate-partner-violence-in-uganda-kenya-and-tanzania-summary
A Framework for Assessing Effects of the Food System http://www.nap.edu/catalog/18846/a-framework-for-assessing-effects-of-the-food-system
Comments Off on Just when I get busy, a pile of interesting NAP reports
Filed under global health
What is the data source for the recent study involving drinking patters
From: XXX
Sent: Monday, June 29, 2015 12:15 PM
To: Laura Dwyer-Lindgren; Gillian Hansen; Abraham D. Flaxman
Subject: US County Drinking Data
Hello All,
What is the data source for the recent study involving drinking patters in US Counties from 2002-2012?!
Thanks,
Aaron
From: Abraham D. Flaxman
Sent: Tuesday, June 30, 2015 8:22 AM
Subject: RE: US County Drinking Data
Hi Aaron,
Thanks for your interest in this work. Our study is summarized here: http://www.healthdata.org/research-article/drinking-patterns-us-counties-2002-2012 and published in AJPH:
Dwyer-Lindgren L, Flaxman AD, Ng M, Hansen GM, Murray CJL, Mokdad AH. Drinking patterns in US counties from 2002 to 2012. American Journal of Public Health. 2015 April 23; e1-e8. doi:10.2105/AJPH.2014.302313. http://ajph.aphapublications.org/doi/abs/10.2105/AJPH.2014.302313
The data is from the Behavioral Risk Factor Surveillance System (BRFSS), a rolling telephone survey, which collects around 250,000 interviews a year. You can see the results for Marathon County in context in our mapping tool here: http://ihmeuw.org/3gar
I suspect that there are other people wondering this as well… with your permission, I’ll include some of this correspondence on my blog, https://healthyalgorithms.com/
–Abie
Comments Off on What is the data source for the recent study involving drinking patters
Filed under global health
More interesting reading in JAMA Oncology
These oncology docs really have to think about communicating risk and the affective aspects of decision making: http://oncology.jamanetwork.com/article.aspx?articleid=2239094
Next I can read ref 1: http://jnci.oxfordjournals.org/content/103/19/1436
Comments Off on More interesting reading in JAMA Oncology
Filed under health communication
Book’s got a cover!
An Integrative Metaregression Framework for Descriptive Epidemiology – http://www.amazon.com/Integrative-Metaregression-Descriptive-Epidemiology-Publications/dp/0295991844
Filed under disease modeling, Uncategorized
Healthy Algorithms 