84,684 reputation
8138292
bio website quantdec.com
location Northeastern US
age 14
visits member for 3 years, 11 months
seen 5 hours ago

Consultant (environmental and spatial stats a specialty), expert witness, and teacher. I can be reached through (outdated but still valid) links posted on my web site.

Twitter: @WilliamAHuber // ASA-P website: http://amstatphilly.org/


Why waste time learning, when ignorance is instantaneous?

--T(iger) Hobbes.

For any complex problem there is a simple solution. And it's always wrong.

--[Mis?]attributed to H.L. Mencken by Dava Sobel, Longitude.


5h
reviewed Leave Open Standard error of the sampling distribution of the mean
5h
comment Normal distribution to triangular distribution
The affine transformation $x\to (b-a)x + a$ maps the interval $(0,1)$ into $(a,b)$ in a one-to-one fashion (provided $a\ne b$). We don't have to worry about the endpoints.
5h
answered Standard error of the sampling distribution of the mean
6h
comment Proof/Derivation of Residual Sum of Squares (Based on Introduction to Statistical Learning)
I was not pointing to a simplification, but to a distinction: the expectation of the square does not equal the square of the expectation. Even after the edits your question does not seem to recognize this crucial fact.
6h
comment Normal distribution to triangular distribution
I do not understand that last comment about closed intervals. Since the uniform distribution assigns no probability to the endpoints $\{0,1\}$ it would seem to make no difference whether the interval is open or closed.
9h
comment Normal distribution to triangular distribution
@Cool If you just want to match moments, use $(v+\sqrt{6})\mu/v$ for the upper limit and $\mu-\mu\sqrt{6}/v$ for the lower limit of the triangular distribution where $\mu$ is the Normal mean and $v$ is its CV. (Whether this makes any sense or not in the OP's application is entirely another matter.)
10h
comment Calculating confidence intervals when binary input variable equals zero
One problem is that you are trying to use confidence intervals where you need a prediction interval procedure. Another is that you seem not to be accounting for the error in the regression estimates. A third potentially is that you are not accounting for the possibility of correlation among the values of the profile variables. A fourth is that you seem to be conflating accuracy with precision: when you neglect sources of uncertainty you get more precise intervals but in fact they probably have much less coverage than you planned on, making them less accurate.
10h
comment What do vertical bars mean in statistical distributions?
Context: you know what the variables mean. When the stuff to the right of the bar is not a random variable, then the bar is not denoting a conditional distribution: it is merely a shorthand for a parameter of a distribution.
10h
answered Follow-up question: When should you center your data & when should you standardize?
10h
comment What do vertical bars mean in statistical distributions?
It would clarify matters to define what you mean by a "conditional distribution" and explicitly distinguish it from a "conditional probability." The uses of "$|$" in the first and third lines are really quite different, despite the similarity of notation, so relying on the very same word "conditional" to distinguish them might be more confusing that enlightening.
11h
comment Proof/Derivation of Residual Sum of Squares (Based on Introduction to Statistical Learning)
Your notation is mystifying because $\mathrm{E}(Y - \hat{Y})^2 = \mathrm{E}[f(X) + \epsilon - \hat{f}(X)]^2$ literally means the square of the expectation. Assuming $\mathrm{E}(\epsilon)=0$, this immediately reduces to $(f(X)-\hat{f}(X)+\mathrm{E}(\epsilon))^2$ = $(f(X)-\hat{f}(X))^2$. Evidently, then, what you really want to compute is the expectation of the square, $\mathrm{E}[(f(X)-\hat{f}(X)+\epsilon)^2]$. But if so, the very first step in your derivation makes no sense. Could you edit the question to clear this up?
1d
comment R rpart classification tree error
This question appears to be off-topic because it is only about debugging R. It is not in good enough shape to be migrated to SO (where it belongs), because a reproducible example of the behavior is required.
1d
comment Represents istances with multiple values for an attribute and similarity between them
I am afraid that does not clarify what you would like to do. What is the relationship between a "user" and the movie information given in the question? What is the meaning of the intended "similarity"?
1d
comment List of situations where a Bayesian approach is simpler, more practical, or more convenient
Thank you sharing your experience with us. Welcome to our site!
1d
reviewed Approve suggested edit on Generalized log likelihood ratio test for non-nested models
1d
reviewed Reviewed List of situations where a Bayesian approach is simpler, more practical, or more convenient
1d
reviewed Close CV.glmnet results
1d
comment Compare linear regression slopes for two datasets, each with a different predictor variable
If the $X_i$ are really six orders of magnitude apart, it sounds like they are different things altogether or at least have been expressed in different units of measurement. Wouldn't that make the slopes automatically uncomparable? (2) Why are you not just doing a multiple regression of $Y$ versus $X_1$ and $X_2$? Is it (perhaps) because you have no data for which $X_1$ and $X_2$ are simultaneously available? (3) Exactly what F-test are you referring to?
1d
comment Represents istances with multiple values for an attribute and similarity between them
What exactly is your "use case"? What are you trying to learn about these data?
1d
comment How to perform a meta regression with a random effect model?Which model should I use?How to start? (beginner)
Thank you for your thoughtful comments and the references you provide. Welcome to our site!