Reputation
12,409
Next tag badge:
99/100 score
12/20 answers
Badges
4 38 91
Newest
 Enlightened
Impact
~471k people reached

Mar
30
revised Best practice for international Fama-French analysis
added 39 characters in body
Mar
30
asked Best practice for international Fama-French analysis
Mar
24
awarded  Enlightened
Mar
24
awarded  Nice Answer
Mar
23
awarded  Nice Answer
Mar
2
comment Why do Human traders make money?
If any of the answers were helpful it is considered good practice here to accept it - Thank you.
Mar
2
revised Time-series similarity measures
deleted 7 characters in body
Feb
29
comment Why do institutions backtest?
"Mostly because of convention and tradition." - I think this is wrong, please see my answer.
Feb
29
answered Why do institutions backtest?
Feb
26
comment Return Attribution: Possible remedies for multicollinearity
I think there is a tradeoff: You are much more flexible with constrained regression but you have to know exactly what you are doing, with ridge regression the process is more automatic so it is in a way less dangerous.
Feb
26
comment Return Attribution: Possible remedies for multicollinearity
Well, removing the extra degree of freedom by adding constraints to certain variables seems a little forced (and certainly not "standard"). Ridge regression is a far better alternative to reduce the variance in my opinion. I am curious if you find something on the net.
Feb
26
comment Return Attribution: Possible remedies for multicollinearity
Although constrained regression can be useful in some circumstances I fail to see how this can address this specific problem of multicollinearity. Can you give a source? (I didn't find any references concerning this point in the documents you cited). And could you please also give a source for your claim that it is the "industry standard for addressing this problem"? Thank you
Feb
26
comment How to calculate cumulative returns with one lag in R
what have you tried so far? what have you achieved? what is the exact problem?
Feb
26
revised Return Attribution: Possible remedies for multicollinearity
added 49 characters in body
Feb
26
comment Return Attribution: Possible remedies for multicollinearity
The problem with PCA is that in this case the new variables won't be interpretable any more (see also point 3 in my answer).
Feb
26
revised Return Attribution: Possible remedies for multicollinearity
edited tags
Feb
26
answered Return Attribution: Possible remedies for multicollinearity
Feb
25
awarded  Nice Question
Feb
23
comment Density plot of the skew-t distribution
It would be helpful if you could post the resulting plots here.
Feb
22
revised How is this financial product called?
added 1 character in body