12,421 reputation
23567
bio website fromthebottomoftheheap.net
location Regina, Canada
age 38
visits member for 4 years, 7 months
seen 6 hours ago

I'm Quantitative Environmental Scientist in the Institute of Environmental Change & Society, at the University of Regina, Canada. I undertake research on environmental problems, including climate change and atmospheric pollution, affecting lakes. I use lake sediments to look back in time at the history of lakes to look at what organisms are present and how the species in the lake have changed through time and how lakes evolve and respond to pollution and perturbations.

I'm also an Adjunct Professor in the Department of Biology at the University of Regina.


17h
comment What is the baseline level in a factor-by-factor interaction?
Yes, year1998 is the modification you need to make if you are in 1998; all samples in 1998 take this modification. The colourWhite:year1998 is an extra modification that you take only if you are that specific combination. If you look at the model matrix (a shown in my Answer) the parametrisation is essentially a set of indicator variables. If you look closely at rows in the model matrix that correspond to rows in the data with colour == "White" & year == "1998" you'll see that all such entries have a 1 in the year1988 column.
1d
comment What is the baseline level in a factor-by-factor interaction?
For your second comment, yes; if you are colourWhite in 1997 then you take the intercept plus colourWhite coefs. If you are colourWhite in 1998 you want intercept, colourWhite, year1998, & colourWhite:year1998 coefficients to get the mean for that combination.
1d
comment What is the baseline level in a factor-by-factor interaction?
"What about comparisons not directly shown in the regression table?" That table just shows the parameters of the particular model parameterisation you have used. To get the comparison you want, you need post hoc comparisons. The multcomp package is particularly strong on this. effects is good too and whilst it generates the effects displays it does also provide the statistics behind the plots.
2d
reviewed Close Post-estimation tests for ordinal probit
2d
answered What is the baseline level in a factor-by-factor interaction?
Apr
16
comment Cross validated $R^2$ and the adjusted $R^2$?
Your Answer doesn't really address the question of the OP, which is what are differences/similarities between a cross-validated $R^2$ and the adjusted $R^2$.
Apr
12
awarded  Cleanup
Apr
12
revised How to include an interaction term in GAM?
wanted indentation that way to compare the s terms
Apr
12
revised How to include an interaction term in GAM?
rolled back to a previous revision
Apr
9
answered Model selection and comparison in GAMM using R (mgcv)
Apr
8
comment What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?
@whuber or am I missing that k-means minimising the sum of Euclidean distances from cluster points to cluster centroids is equivalent to minimising the Euclidean distance between cluster points and the medoid?
Apr
8
comment What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?
Right, but k-medoids with Euclidean distance and k-means would be different clustering methods. I don't see the OP mention k-means at all. The Wikipedia page you link to specifically mentions k-medoids, as implemented in the PAM algorithm, as using inter alia Manhattan or Euclidean distances. The OP's question is about why one might use Manhattan distances over Euclidean distance in k-medoids to measure the distance to the current medoids. Hence I don't understand what you think is wrong with the OPs question?
Apr
8
comment What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?
Can you provide a link to the Wikipedia page that says the "algorithm is not defined for Euclidean distance...". The page @whuber links mentions that the Euclidean distance can be used to measure distances to medoids, and that the method can work with an arbitrary matrix of distances.
Apr
8
reviewed Reviewed nnet function in R
Apr
8
revised nnet function in R
added 196 characters in body
Apr
8
reviewed Edit What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?
Apr
8
revised What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?
deleted 31 characters in body; typo in title
Apr
8
comment What is the benefit of using Manhattan distance for K-medoid than using Euclidean distance?
@whuber The page you link to gives a different distinction between k-mediods and k-means. The former uses mediods whilst the latter uses centroids. The algorithm that the page describes (PAM), states that any valid distance may be used in PAM to measure the distance between the observations and the current mediods, and gives the Euclidean distance as one such choice. As such, I find your comment difficult to "make sense of", at least in terms of you pointing the OP to a resources than seems contradictory to your observation.
Apr
8
reviewed Close Matlab FACTORAN error on line 162: a covariance matrix is not positive definite
Apr
2
answered interpreting NMDS ordinations that show both samples and species