Reputation
10,675
Top tag
Next privilege 15,000 Rep.
Protect questions
Badges
24 57
Impact
~636k people reached

Apr
27
revised “Unidentified” hierarchical model in brms/stan - where to go from here?
added 62 characters in body
Apr
27
answered “Unidentified” hierarchical model in brms/stan - where to go from here?
Apr
17
comment Accelerometer normalization
So you want to normalize the range as well as the change over time? So all the values end up with something like a range of [0, 1] and complete their change over 500 time units, for example? What's the goal of this normalization? (Graphing, modeling, etc.)
Apr
6
reviewed Reviewed t Test, Chi Squared or logistic regression..?
Apr
6
comment Deterministic clustering approaches
I agree with Tim that knowing a bit more about the problem would help. When you say "deterministic" do you mean the same things end up in clusters together each time, or that they end up in exactly the same cluster number? Is the number of clusters fixed or not? When you say "predict their behavior" are you really trying to come up with a supervised method that uses cluster numbers as its target? Or something else?
Mar
31
awarded  Popular Question
Mar
29
revised How to predict orders when unit of measure is different across and within months
added 327 characters in body
Mar
29
answered How to predict orders when unit of measure is different across and within months
Mar
29
comment Need to extrapolate missing monthly data from annual data; the monthly/seasonal index is reasonably well-known
Whatever approaches are suggested, you should obviously hold out some of the monthly-revenue companies as test cases and see what happens when you impute monthly values from their annual totals.
Mar
28
awarded  Custodian
Mar
28
reviewed Reviewed Linear equivalent to chord diagram
Mar
28
reviewed No Action Needed What is text distance in data mining
Mar
27
revised Why use vector error correction model?
added 67 characters in body
Mar
26
comment Measure of central tendency for periodic variable (hour of day)
Look under the name in the tag: "circular statistics". There's an R package for that, for example.
Mar
25
revised What to do if time series data remains autocorrelated?
added 564 characters in body
Mar
25
revised What to do if time series data remains autocorrelated?
deleted 281 characters in body
Mar
25
revised What to do if time series data remains autocorrelated?
added 191 characters in body
Mar
25
revised What to do if time series data remains autocorrelated?
added 24 characters in body
Mar
25
answered What to do if time series data remains autocorrelated?
Mar
23
reviewed No Action Needed I know the probability of an event to occur, so how do I find how many times the event must be attempted to get a reasonably chance of success?