Reputation
35,683
Next tag badge:
389/400 score
110/80 answers
Badges
1 62 138
Newest
 Enlightened
Impact
~2.0m people reached

4h
comment R: Quadratic Regression with interaction: when to center?
What makes you think that such 1 d.f. tests are unambiguous? It is precisely because they are ambiguous (especially the test of the linear term is meaningless) that I recommend against it. Academic settings have the same needs as you on this point. Also, testing the nonlinear term and changing the model ruins $p$-values and confidence intervals. The best result comes from just looking at the plotted quadratic fit, and its confidence band (even better: simultaneous confidence bands).
11h
revised How to get Cox & Snell, Nagelkerke R-Square in R logistic regression output?
added 7 characters in body
12h
answered How to get Cox & Snell, Nagelkerke R-Square in R logistic regression output?
1d
revised Does the p value for logistic regression depends on odds ratio or logit?
edited tags
1d
answered Does the p value for logistic regression depends on odds ratio or logit?
2d
comment Matching data before regression (multiple treatment variables)
Matching no more deals with causal effects than regression does. The only difference is that sometimes with propensity scores you can handle a larger number of adjustment variables.
2d
comment R: Quadratic Regression with interaction: when to center?
A chunk test is just a contrast consisting of more than one effect.
2d
comment R: Quadratic Regression with interaction: when to center?
It is fruitless to try to find out which one is significant, and besides creating a multiplicity problem and inflating type I error, further testing is unreliable. You can get all the inference you need without that other step.
2d
revised R: Quadratic Regression with interaction: when to center?
edited tags
2d
comment R: Quadratic Regression with interaction: when to center?
What you described is an interaction between one simple variable and a 2-column variable; you need to interact the simple variable with both linear and quadratic terms and summarize evidence for interaction (shape change) with a 2 d.f. "chunk" test.
2d
comment Convert hazards ratio to odds ratio
But RR $\neq$ hazard ratio.
2d
comment R: Quadratic Regression with interaction: when to center?
Centering gets in the way of understanding the model, and doesn't help anyway. Centering doesn't affect predicted values from the model, and tests of effect combine linear + quadratic terms (2 d.f. test) which is unchanged by centering. Same for interaction effect (2 d.f., interact a variable with both linear and quadratic terms).
Apr
28
revised Logistic regression coefficient too high - cannot interpret odds ratio
edited tags
Apr
28
answered Logistic regression coefficient too high - cannot interpret odds ratio
Apr
28
comment Why Type III ANOVA is used for this analysis of coefficients
Type III contrasts, which are maximum variance contrasts in some situations, are generally discouraged. See the classic discussion by Bill Venables at sedarweb.org/docs/wsupp/S10RD01.pdf.
Apr
28
answered What should be validation strategy?
Apr
28
comment Classification accuracy increasing while overfitting
Though not the dominant effect here, your use of an improper accuracy scoring rule will have a negative impact on what you are trying to do. For example, you can increase proportion "classified" "correctly" by dropping very important features.
Apr
28
comment Data reduction by maintaining data distribution
This is related to William Dumouchel's "Data Squashing" algorithm.
Apr
28
comment Performance of Logistic Regression with time
Alternatively you can fit time as a flexible nonlinear effect and include all the data.
Apr
26
revised Is it normal for logistic regression, to have predictors which have good Wald's Chi Sq, but still bad performance?
edited tags