Reputation
Next tag badge:
972/1000 score
379/200 answers
Badges
8 54 127
Newest
 Nice Answer
Impact
~2.4m people reached

12h
answered Moments of linear combinations of normally distributed variables
23h
reviewed Close How to change x axis to handle outliers
23h
reviewed Leave Open State the distribution of the time he has to wait for four customers to arrive. Sketch its density and calculate mean and variance of the waiting time
23h
reviewed Close Neural network using Matlab
23h
reviewed Close AR-GARCH model with r
23h
reviewed Looks OK What's the name for this statistical fallacy?
23h
answered Why can multicollinearity be a problem for logistic regression?
1d
comment Should I use logistic mixed effects? How?
That sounds right to me
2d
comment P-value versus Exp(B) value in Cox regression analysis
1. Yes, I favor hazard ratios over p-values. 2. The wide CI on the second biomarker is a bit worrying and is the reason that it has a higher HR with a less sig. result. 3. both p are very sig., and I believe estimates of p values that are very low are not always that accurate (but I'm not sure of that)
2d
reviewed Leave Closed Statistical model of discrete function of multiple parameters
2d
reviewed Reopen Correlation of motion or movement timeseries
2d
answered P-value versus Exp(B) value in Cox regression analysis
2d
answered Should I use logistic mixed effects? How?
Aug
29
reviewed Approve Variance and autocorrelation with missing and/or unevenly spaced data in time series
Aug
29
comment Model selection, issues of judgement
Well, there is always @FrankHarrell 's book Regression Modeling Strategies.
Aug
29
comment Difference between modelling and testing association
The language isn't always consistently used. It does get confusing. A "test of association" refers to e.g. correlation and chi-square. However, the results of a regression are often described as association. This is mostly to avoid causal language. It might be better to use "related" for one and "associated" for the other ... but it will be confusing, regardless. The main thing is that if you have a dependent variable, you probably want some form of regression.
Aug
29
comment Difference between modelling and testing association
If you want to test association, use a test of association. But then you don't have an outcome variable. If you have an outcome variable, use some form of regression. After you do logistic, you don't need to do ANOVA or t-test (and, in fact, ANOVA is inappropriate if the dependent variable is a dichotomy.
Aug
29
comment Model selection, issues of judgement
You will need to have a large enough N to avoid overfitting. If you have 10 variables and want to look at all (10*9)/2 = 45 interactions that's 55 independent variables, which is a lot unless you have a pretty large data set. Exactly what to do will depend on the exact situation
Aug
29
answered Model selection, issues of judgement
Aug
29
reviewed Close How to determine whether 2 code snippets are functionally same?