Jonathan Lisic
Reputation
Top tag
Next privilege 1,000 Rep.
 Apr 21 comment Should potentially multicollinear variables be dropped Is there any meaning behind these independent variables? If the underlying theory that you are working with is based on these variables I would keep them. If you are doing model selection, I would just look at the performance of your model with and without the variables, take a look at the residuals and see how well the model fits. Also check the coefficients in your model to check for any tell-tale signs of multicollinearity such as large coefficients with alternating signs. Mar 12 comment How to calculate MLE of variance? The variance is an attribute of a random variable, what random variable are you interested in finding the variance for? Maximum Likelihood Estimates, are estimates of parameters. In the case of a binomial distribution, you have a single parameter $\theta$. The estimator for $\theta$, $\hat{\theta}$ has a variance. Feb 17 revised Slope of regression line in Moran scatterplot Minor edit in title, fixed references so that they are readable. Feb 17 comment Slope of regression line in Moran scatterplot Can you clarify your question? "Is it true?" what is it? Feb 17 suggested approved edit on Slope of regression line in Moran scatterplot Feb 17 comment Types of Mean in statistics Is this a homework problem? All of these means are described in wikipedia, the interquartile and trimean are both ways to decrease the impact of extreme values in a measure of central tendency, "mean." Jan 26 comment If my goal is to show very low correlation, how should I check for statistical significance? Are these two variables normally distributed? Jan 25 comment Probit Model: Interpretation of marginal effects if explanatory variables are proportions Sorry, I misread the original question, @Fuca26 is entirely correct. It is important to note that the marginal effect is a linearization, and only meaningful around the current parameter value. Jan 25 revised How high dimensional t-test in Feng's article has been obtained? Appropriate Titles for Authors Jan 25 comment How high dimensional t-test in Feng's article has been obtained? I think you need to clarify what your specific question is, what do you understand? what don't you understand? Jan 25 suggested approved edit on How high dimensional t-test in Feng's article has been obtained? Dec 16 comment Gradient Boosting for Linear Regression - why does it not work? I like the answer, but to be a bit pedantic, $\beta$ from regression is the best linear unbiased estimator. Dropping unbiasedness may allow you to do a bit better particularly under high multicollinearity, something you eluded to at the end. Dec 15 comment How do I compare multiple runs of K-means? The Rand index and the adjusted rand index provide a classical approach, this statistic is simply based on pairwise counts between clusterings. en.wikipedia.org/wiki/Rand_index Sep 5 revised Discretizing a Continuous Input for an Artificial Neural Network ANN was stated several times without defining, the neural-network tag was specified but it is unclear how much this will help with google searching. Sep 5 suggested approved edit on Discretizing a Continuous Input for an Artificial Neural Network Aug 16 comment Is there a multiple testing problem when performing t-tests for multiple coeffcients in linear regression? Sorry I don't have time right now to give you a better answer, but the results from the regression are answering a particular hypothesis test (comparing distributions under the inclusion or removal of a particular variable. A Bonferroni-like adjustment would instead be more appropriate if the null hypothesis was more complicated including a vector of parameter. This is the case in multiple comparisons. Jun 9 comment Fitting the differences between two curves Jun 9 comment Fitting the differences between two curves I'm still not happy with the definition; maybe a better statement would be along the lines of testing that the difference between two 'lines' are decreasing over a fixed interval. So what is generating these 'lines'? Jun 9 comment Fitting the differences between two curves I think you really need to flesh out what convergence means, in explicit language. If two things cross and remain close they can be parallel for a sufficiently small difference between each other, you could also have to coincident vertical lines, they clearly converge to each other, but diverge in several definitions of convergence. Is anything here random? Jun 9 comment Fitting the differences between two curves Can you define converge and diverge in a slightly less visual sense, and what the properties of these curves are? Remember convergence is defined as one object becoming another object in some sense such as the convergence of a sequence of values to another value, or convergence in distribution of a random variable to another random variable.