336 reputation
27
bio website
location Aylesbury Buckinghamshire, UK
age 30
visits member for 2 years, 8 months
seen 45 mins ago

quadratic linear form XFX with matrix

F:  1   1
    1   1

in canonical base can be expressed as X' F' X'

X'  1   0   X'   
    0   0

where X'=A-1X, F'=AFA and A is transition matrix. Canonical base vectors are [1,0] and [-1,1]. Thus this is correct that (a+b)^2 = a^2 + b^2 + 2ab


Aug
24
revised Pairs trading: Question on non-negative profits, size of the positions and trading signals
refactor
May
5
comment Does DOM trading using broker data make any sense?
thank you very much
May
4
comment Interpolate option volatility in delta space in R
interpolation, linear or cubic or any else between two delas that are implied, i.e. fit to BS, with respect to time, or strike doesn't yield deltas from BS, because none of these variables is linear function of delta. so ideally you want to get max close to each other real implied deltas and interpolate between them what minimizes the errors
May
4
comment Interpolate option volatility in delta space in R
moreover you know that this spline will introduce errors: deltas between nodes will not fit into BS
Apr
28
comment Time Varying Volatility
yes, then this is true
Apr
28
comment Time Varying Volatility
yes, then this is true
Apr
28
comment Time Varying Volatility
no, again, not. note: you can have positive and negative autocorrelation
Apr
28
comment Time Varying Volatility
no, again, not. note: you can have positive and negative autocorrelation
Apr
28
comment Time Varying Volatility
no, this is not true
Apr
28
comment Time Varying Volatility
no, this is not true. this is very possible that then there is heteroskedacity but it is not invariant
Apr
26
revised Time Varying Volatility
added 437 characters in body
Apr
26
answered How does one use the Johansen cointegration test in a linear time series model?
Apr
26
answered Time Varying Volatility
Apr
17
comment Using cointegration to prove that a long-short strategy is market neutral (in CAPM sense)
you will get this if you note your PL as PL=y-bx and then you want to minimize the variance of this PL, you get beta
Apr
17
comment Using cointegration to prove that a long-short strategy is market neutral (in CAPM sense)
yes it is: because errors e=y-bx are stationary (at least in a weak sense)
Apr
17
comment Using cointegration to prove that a long-short strategy is market neutral (in CAPM sense)
@GoodGuyMike no market neutrality in cointegration by definition, cointegration in definition has nothing related to "Market Neutrality", all cointegration tells us is that errors will be nice, which is also what we are interested in when searching for neutrality
Apr
16
answered Using cointegration to prove that a long-short strategy is market neutral (in CAPM sense)
Apr
6
revised Using variance ratios to test for mean reversion
added 81 characters in body
Apr
6
revised Using variance ratios to test for mean reversion
added 74 characters in body
Apr
5
answered Using variance ratios to test for mean reversion