# highBandWidth

less info
reputation
211
bio website location age member for 2 years, 11 months seen Nov 7 at 22:27 profile views 66

# 137 Actions

 Sep18 comment Area under the “pdf” in kernel density estimation in R @Anony-Mousse, yes, that is what this question is asking. Why is it not evaluating to 1? Jan25 awarded Popular Question Jan10 awarded Yearling Mar20 comment Numerical or computational method to generate density estimate @Henry, I guess by complicated I mean that it may be less smooth in some parts, say something like$$Y[x]=\sin \frac{10}{(x+10^{-3})^2}$$ I guess I am just coming up with something that needs more samples in certain parts than others. Please don't find fault with the exact function. Mar20 asked Numerical or computational method to generate density estimate Mar5 asked Public databases of learned HMM models for NLP Feb29 asked Statistical model of a website Feb21 answered Distribution of execution time for benchmarking? Jan29 accepted Interpolating between models in ROC space Jan29 asked Consistent ranked list for ROC interpolation Jan28 comment Formulating and fitting a nonlinear model that looks like a product of linear models Why do you think this would be the best model? I am asking because the same line of thinking would probably lead to a good way of inferring the $\beta$s. Jan28 comment Convergence of identically distributed normal random variables What does it mean to be identical copied of $X_1$? Is $X_2=X_3=...=X_n=X_1$? How are they dependent? To be dependent would mean $P(X_1,X_i) \neq P(X_1) P(X_i) \forall i \in [2,n]$. What then is $P(X_1,X_i)$? Jan26 comment Accounting for discrete or binary parameters in Bayesian information criterion I guess my question was, is the DIC an approximation of the marginal model likelihood too? I guess I should read about it myself, but since we were discussing it, I thought explaining this would make the answer more complete. Thanks! Jan25 answered Interpolating between models in ROC space Jan25 answered Normalization in pairwise hypothesis testing Jan25 comment Accounting for discrete or binary parameters in Bayesian information criterion So I am slightly confused. I thought BIC was an approximation of $E[log P(y|Model)] = \log(\int P(y|\theta)P_{model}(\theta)d\theta)$, which can be calculated from the MCMC simulations. Why then would we calculate DIC? Jan25 asked Interpolating between models in ROC space Jan25 comment Optimising for Precision-Recall curves under class imbalance @JackTanner, The model that I have (MARS with logit function) gives outputs in the range of 0 to 1, similar to logistic regression. It's basically the same, but includes a few more features. To get precision at different recalls, I simply set the thresholds at different points. I just use the standard way to calculate PR or ROC from a ranked list. Jan24 comment Optimising for Precision-Recall curves under class imbalance @JackTanner, everything is computed on the training set for now. Since the model does not have that many parameters, and the number of samples in the training set is huge I don't worry too much about overfitting. Besides, I want to be sure I am getting good performance on the training set before I can expect in in the test set. Jan24 revised Optimising for Precision-Recall curves under class imbalance clarified the different models and evaluations