Reputation
15,067
Top tag
Next privilege 20,000 Rep.
Access 'trusted user' tools
Badges
2 39 75
Newest
 Nice Answer
Impact
~918k people reached

19h
comment add legend to ggplot geom points
You've fundamentally misunderstood how ggplot works. I'm assuming you have female bats? If so, do ggplot(subset(random_df, Sex == "male"), aes(x = speed, y = TURNANGLE2, colour = activity)) + geom_point(). Or at least that's what i think you need, without seeing a reproducible example. This really ought to have gone to Stack Overflow, for future reference.
19h
revised add legend to ggplot geom points
added 208 characters in body
Feb
7
comment Logistic regression model does not converge
By convergence I mean that the parameters being estimated in the model don't change (or only change less than some small tolerance) between iterations. Here the parameters get increasingly large and fitting stops because of the limit on iterations but the parameter estimates changed a lot between the penultimate and the last iterations and as such haven't converged.
Jan
27
revised Conditional Mean in Linear Regression
edited title
Jan
26
awarded  Nice Answer
Jan
25
revised How to include an interaction term in GAM?
added 28 characters in body
Jan
25
comment How to include an interaction term in GAM?
No, the first smooth is the global function and the by smooth represent site-specific differences between it and the global smooth. The by smooths really need m = 1 added to them to put the penalty on the first derivative for the difference smooths though.
Jan
22
comment Smaller p-values on a selective multiple regression analysis
I'd look at the magnitude of the estimated effect (the coefficient) and the corresponding standard error for each of the six terms in the reduced (selected) model and the "original" p value. Both will have changed, and in particular the standard errors will tend to be smaller; this is the bias variance trade-off at work. Anyway, those p values you quote are conditional upon the other terms in the model; change the other terms in the model, change the p value.
Jan
22
revised showing the autocorrelation fn of a moving avg q MA(q) time series
Tried to add some math markup - not right yet as I wasn't exactly sure of the OP's intentions
Jan
20
revised How do we put various multivariable data in cluster bucket
added 406 characters in body
Jan
20
comment How do we put various multivariable data in cluster bucket
Once you have the distance between each site the clustering proceeds as any standard hierarchical cluster analysis, that's what agnes() does. The type of data/variables is irrelevant once you have the distances. Is your question more about what is hierarchical clustering? If so, that's not clear from your question.
Jan
20
answered How do we put various multivariable data in cluster bucket
Jan
20
answered Presence/Absence GAMM with penalized smoother convergence/completion problem, data set too large
Jan
20
answered Can I use bootstrapping to estimate the uncertainty in a maximum value of a GAM?
Jan
18
answered How I can interpret GAM results?
Jan
17
revised Gibbs sampler gets stuck in local mode
edited title
Jan
5
answered Predicting mean smooth in GAM with smooth-by-random-factor interaction
Jan
5
comment Predicting mean smooth in GAM with smooth-by-random-factor interaction
@jaimedash no (but plogis(response)), I don't think so (i.e. yes I think this is all you need). You're supposed to do all these things on the scale of the linear predictor and then back transform. More generally you can always grab the inverse of the link function via ilink <- family(mod)$linkinv and use ilink(foo).
Jan
4
comment GLM with logit link and Gaussian family?
@MaartenBuis Indeed; I didn't intend this to be taken as an either/or - I've also used both quasi-binomial and beta regressions.
Jan
4
answered GLM with logit link and Gaussian family?