6,140 reputation
1334
bio website integrativestatistics.com
location Boston, MA
age
visits member for 3 years, 10 months
seen 5 hours ago

Nov
16
answered Factor loading and corrected item to factor correlation
Nov
14
awarded  Popular Question
Nov
12
comment Need Insight on Measuring Variable Effect
Consider the possibility that what you need is not an answer from some knowledgeable person who's at arm's length from the work, but either a consultant or a coach who can get to know in detail the data, your analyses thus far, and your level of understanding and analytic capabilities.
Nov
1
awarded  Tag Editor
Nov
1
revised heteroscedasticity wiki excerpt
Heteroscedasticity is a property, not a process.
Nov
1
comment How to test heteroskedasticity at the independent variable level?
What method would you use in the case of residuals, and why would that not apply for a predictor?
Nov
1
suggested suggested edit on heteroscedasticity tag wiki excerpt
Oct
31
comment Sample size calculation for non-normal data (possibly lognormal)
You seem to be saying that based on three observations/cases/data points/"numbers" you have determined the shape of the variable's distribution. And that using a power calculation you have determined that in order to achieve a certain level of power you will require not 3 but 2 cases.
Oct
31
comment Statistical analysis of variable data
Needs to be clarified and reorganized.
Oct
23
revised Best practice for visualizing a lift from zero?
edited tags
Oct
23
comment Best practice for visualizing a lift from zero?
I would say that what depends is not the limit of the ratio but one's interpretation of that ratio.
Oct
23
comment Number of possible statistical test can be used on the observation data.
Please specify what you mean by "show the relationship between the known and unknown factors."
Oct
18
comment How to make a vertical line in R
Question seems on-topic as it is about data visualization. I'd like to see a fuller explanation for the effective code described by @user777.
Oct
14
awarded  Popular Question
Oct
10
comment Prediction when survey subsets create dramatically smaller Ns
Every person has a Y value.
Oct
7
asked Prediction when survey subsets create dramatically smaller Ns
Oct
3
comment What kind of test is this?
The method you describe has a bit in common with what's sometimes called monadic testing. Sources that go into that might also go into your method.
Oct
3
comment How to apply a logistic + an OLS model to the same data set?
Possibly censored regression would apply. I learned a lot from stats.stackexchange.com/questions/49443/…
Oct
1
comment Multiple Linear Regression: Obtaining a Stable Model
Remember that under random sampling a relationship with p=.05 has a 50-50 chance of producing a p>.05 among the next sample chosen. Sound like you've overfitted.
Sep
30
awarded  Explainer