8,540 reputation
31540
bio website
location England
age
visits member for 1 year, 4 months
seen 1 hour ago

Apr
17
revised Can I write estimate $\pm$ its standard error?
edited tags
Apr
17
answered Can I write estimate $\pm$ its standard error?
Apr
17
comment Rigorous Bell-CHSH?
Van Dam et al. (2003) might be of interest.
Apr
17
comment How to interpret Weka logistic regression output for a nominal attribute and its coefficients?
It's arbitrary - your software might assign codes by alphabetical order, by the order the levels are first encountered in the dataset, or by the order given when you define the levels.
Apr
17
reviewed Close PCA on non-centered data
Apr
17
comment How to interpret Weka logistic regression output for a nominal attribute and its coefficients?
If it's not in the manual or wiki try asking the mailing list. Or if you've a general question about how to code binary predictors please try to clarify it.
Apr
17
reviewed Close How to interpret Weka logistic regression output for a nominal attribute and its coefficients?
Apr
17
comment How to interpret Weka logistic regression output for a nominal attribute and its coefficients?
This question appears to be off-topic because it is about the detail of how Weka codes predictors.
Apr
17
comment How can I improve the predictive power of this logistic regression model?
... regression models with other classification techniques that don't provide a predicted probability.
Apr
17
comment How can I improve the predictive power of this logistic regression model?
@user1205901: It's silly unless the sole purpose of the model is to decide between alternative courses of action & the cost of making the wrong decision is the same for each. (If the alternatives were to carry out a blood test or to do nothing; & a patient's symptoms indicated a 49% chance of his having cancer, would you really send him home?) Even then, a model that predicts a 55% chance of cancer in someone who's in fact healthy is unlucky; a model that predicts a 99.99% chance is suspect. Perhaps the main reason to use classification accuracy is to allow comparison of logistic ...
Apr
17
answered Differences in edition of advanced statistics books
Apr
16
comment Is it ok to spit non-normal variables in tertiles and put them into multivariate regression models?
Not a very good idea - see here.
Apr
16
comment How can I improve the predictive power of this logistic regression model?
For a start use a proper scoring rule to assess model performance, unless you can justify that 0.5 cut-off.
Apr
15
comment What is PRINCIPAL HESSAIN Direction Model, how and where can I use it?
As far I know principal Hessian directions are used for post-hoc data reduction & not tied to any particular experimental design. Don't think anyone can tell you what kind of design to use for your experiment without your giving more specific information about the problem. And there are whole books written on regression - exactly what trouble are you having? ("Conical analysis" I've never heard of.)
Apr
15
comment tribal food data analysis
Why is it a problem that some variables have non-normal distributions? - What analysis do you want to perform?
Apr
15
reviewed Close metatrim command version 11
Apr
14
comment Treating ordinal variables as continuous for regression problems
Re @Frank's suggestion of penalizing predictors towards ordinality: see the reference given here
Apr
14
reviewed Approve suggested edit on Choosing a regression model based on missing values
Apr
14
awarded  Fanatic
Apr
14
reviewed Approve suggested edit on I am unsure how to execute and interpret a simple linear regression model where Y ~ a single X