2,032 reputation
426
bio website
location Germany
age 27
visits member for 1 year, 8 months
seen Nov 15 at 13:19

Nov
6
reviewed Approve suggested edit on Show that n is a perfect square if and only if $k_i$ is even for $ 1 \leq i \leq m$
Nov
6
comment Estimating parameters for a binomial
I recommend posting the question on stats/crossvalidated SE.
Nov
6
comment Variances for K-Means clustering
It would be helpful if you could tell us what the variables stand for.
Nov
6
answered Transforming frequency data into a 'rating' or 'valuation'
Nov
6
answered Statistic t table
Nov
6
comment I want to prove Determine the coupon rate $r$, such that the price of the bond, at $T_0$, equals its face value.
Is there an independent interest rate here, or is there no discounting? If there is no discounting, then $r=0$ implies the value of the bond is $K$ at $t=0$.
Nov
6
comment Transforming frequency data into a 'rating' or 'valuation'
So why not rank your items according to frequency? Rank 1 to item with lowest frequency, rank 2 to the one with second highest frequency etc? You only have to keep one row per item to keep track of the cumulative frequencies, to be updated as new data comes in. For better answers you probably have to give us more information. In particular, why should a ranking based on frequency alone correspond to a ranking based on the value of these items? What is the ranking supposed to achieve: rank according to value or frequency or something else?
Nov
4
comment Problem on EU commission
Here the example; $z$ isn't strictly concave, but at the maximum it is. wolframalpha.com/input/?i=plot+%280.52-0.5%2Bx*%280.53-0.5%29%29%2F%2‌​80.52*0.48%2Bx^2*0.53*0.47%29^%281%2F2%29+for+x%3D[0%2C5]
Nov
4
comment Problem on EU commission
It took me a while, but I am now convinced your $\max F(z)$ and my $\max 1-\phi(-z)$ objectives are identical (assuming $\sum c_i=1$). Exposition could be clearer at this point. ;) But then you look at the gradient and determine the weights - how exactly? $F(z)$ is not strictly concave, so the first order approach to determine the max need not be valid. Is $z$ always strictly concave, as your graph seems to suggest? One example of mine seems to suggest that isn't always so. Then we might have a nasty case of convex optimization. Otherwise your theorem seems to be correct.
Nov
4
reviewed Approve suggested edit on Group theory problem automorfism
Nov
4
reviewed Approve suggested edit on Expected value of inverse of a random variable
Nov
4
answered Problem on EU commission
Nov
4
awarded  Custodian
Nov
3
awarded  Revival
Oct
23
comment Round robin logistic regression
Let's wait and see if somebody else has an actual answer, and not just a quick and dirty trick. ;)
Oct
21
comment Round robin logistic regression
I am not entirely sure if it always adds up to 1. But even if it doesnt, just renormalize: $P(A|X)/[P(A|X)+P(B|X)+P(C|X)+P(D|X)]$.
Oct
20
comment Round robin logistic regression
Why not use a multinomial logit, which explicitly allows for more than two responses? Then probabilities are guaranteed to sum to one.
Oct
20
comment Welford's algorithm for standard deviation: combine multiple sets of results
If you have the standard deviation for each set, you could compute the pooled standard deviation: en.wikipedia.org/wiki/Pooled_variance
Oct
20
comment How to calculate probabilities of win, draw and loss based on the ELO system
Easiest way: Define analogously $E_=$ and $Q_=$ for draws, then $E_A=Q_A/(Q_A+Q_B+Q_=)$ etc.
Oct
20
comment Negative Correlation and Securities?
Just look up the definitions of expected value and standard deviation and apply them!