Reputation
Next tag badge:
953/1000 score
371/200 answers
Badges
8 52 125
Newest
 p-value
Impact
~2.3m people reached

3h
reviewed No Action Needed Detecting Gibbs Sampler convergence with Raftery and Lewis Diagnostic
3h
reviewed Looks OK Nagelkerke value equals 1. Why?
3h
comment Statistical significance of the Lorenz curve (Gini index)
I don't know offhand, but there is a book by Handcock and Morris that should have the answer.
15h
comment How do you do a nonparametric quantile regression in SAS?
Then linearity of effect is meaningless in any case.
1d
reviewed Leave Open Forecasting methods for monthly sales
1d
comment How do you do a nonparametric quantile regression in SAS?
To add polynomials, you can just create them in a data step and then add them to the model. Splines are available in QUANTREG see support.sas.com/documentation/cdl/en/statug/63962/HTML/default/…
1d
comment How do you do a nonparametric quantile regression in SAS?
Yes, a nonlinear relationship would need to be modeled by nonlinear terms - the simplest are polynomials, but splines are also possible.
1d
answered How do you do a nonparametric quantile regression in SAS?
1d
reviewed Leave Open Sequential Singular Value Decomposition
1d
reviewed Close How to generate random numbers correlated to a given dataset in matlab
1d
reviewed Close How can I obtain the Simulations in CausalImpact package?
1d
answered What to do if residuals are not normally distributed?
2d
reviewed Close how to modify box plot diagram
2d
reviewed Leave Open Non normal and nonlinear data
2d
reviewed Close how to enhance box and whisker plot
2d
answered hypothesis test one-sample vs two-sample
2d
comment Non normal and nonlinear data
First, no regression method requires that your data be normally distributed, neither the dependent variable nor the covariates need to be.. Second, I am guessing that you mean "linear relationship between depedent variable and covariate". Third, are you sure you have 2 DVs? Are they related? Fourth, what are your variables and what are you trying to find out?
2d
answered Using lm() with just one variable in R
2d
comment Why Removal of Outliers Does not Affect Confidence Interval
Are you sure you removed them? Look at the df for the two models (with and without the extreme point) and see if they differ. (Not trying to be snarky, this kind of error is easy to make. I've done it).
Jul
27
answered Similarity measure for observations with many variable types