8,011 reputation
2141
bio website thinkinator.com
location Washington DC, United States
age
visits member for 4 years, 5 months
seen 1 hour ago

Data Scientist at ERI, an incredible analytics company where we do everything from fraud detection and visualization, to training and and text mining. (Photo by Peter B.)


Mar
24
awarded  Good Answer
Mar
23
awarded  Caucus
Mar
23
awarded  Constituent
Mar
22
comment Negative real wages and log wages
Are you asking how the log of a number can be negative?
Mar
20
awarded  Popular Question
Mar
18
awarded  Nice Answer
Mar
11
comment Bayesian additive regression trees (BART) for classification analysis of gene expression data
BayesTree is done by the inventors of BART, I believe. But the bartMachine package looks like a better option to me. (The only downside is it requires Java and interfaces to it via the rJava package, which might complicate things if your machine doesn't have Java installed.)
Mar
10
comment Is there any way to estimate my model other than SEM?
What software are you using and what models do you know how to run? Your graph looks like it might fit the SEM (Structural Equation Model, not Simultaneous Equation Model) framework, though how you've drawn it is non-standard. Perhaps if you look at how the example is drawn on Wikipedia and redraw, you'd get a better answer: en.wikipedia.org/wiki/Structural_equation_modeling
Mar
9
answered Using Kalman filters to impute Missing Values in Time Series
Mar
6
comment What are the disadvantages of using Lasso for feature selection?
@BerkU.: I'm not sure. I was thinking more of the number of tests you'd end up doing -- especially in the case that conjectures talks about where $k$ varies -- which could be so large in comparison to the amount of data you have that you end up overfitting even if you try CV or other techniques that might avoid overfitting in more normal circumstances.
Mar
6
comment What are the disadvantages of using Lasso for feature selection?
Not to mention that the brute force approach would require more data and more care to avoid overfitting.
Mar
3
comment How to handle missing data in a small $n$ large $k$ machine learning scenario?
So you're talking about replacing roughly 130 values in your dataset (0.001 x 1000 x 130), and these values are impossibly large, is that correct?
Mar
3
comment How to document a Random Forest result (final model)?
@tnaake: Your code and data on a public source is the best way to do things. Most people don't do that because they view their data as proprietary or they're embarrassed that their code is poor. I don't know of any publications that I could point you to.
Mar
2
answered How to document a Random Forest result (final model)?
Feb
27
comment Approach and example of graph clustering in “R”
I was going to recommend igraph as well, but I'll let AndyW type it up. As comments have said, your example is a tough case, since you show three cliques joined together.
Feb
27
revised Why Multiple Linear Regression cannot be built when p>n?
added 626 characters in body
Feb
27
answered Why Multiple Linear Regression cannot be built when p>n?
Feb
27
comment Why Multiple Linear Regression cannot be built when p>n?
I thought you mentioned MLR and LDA multiple times when I first saw your question, but when I edited it just now, to expand MLR to "Multiple Linear Regression", it wasn't there. I don't think I did anything else. At any rate you may want to further edit it to make your MLR v LDA comment part of the question.
Feb
27
revised Why Multiple Linear Regression cannot be built when p>n?
added 26 characters in body; edited title
Feb
27
comment Why Multiple Linear Regression cannot be built when p>n?
And "MLR" means what?