Reputation
451
Top tag
Next privilege 500 Rep.
Cast close & reopen votes
Badges
3 11
Newest
 Tumbleweed
Impact
~12k people reached

Apr
20
comment Sum of two GARCH(1,1) Models
If sum of tho GARCH processes isn't GARCH, how could we model Index and its components at once ? (say small index with up to 20 stocks), in intraday case for example we would like to quantive how sudden jumps in volatility of components could rise volatility of Index (to opposite jumps could be not visible when we look at Index)
Apr
20
comment What kind of errors arise when I fit ARMA(1,1) to data generated from ARMA(1,1)-GARCH(1,1) process?
@Richard Hardy I'm not sure so I ask here
Jan
24
awarded  Tumbleweed
Jan
16
asked Quantifing quality of cointegration
Oct
22
awarded  Popular Question
Feb
5
revised What kind of errors arise when I fit ARMA(1,1) to data generated from ARMA(1,1)-GARCH(1,1) process?
added 7 characters in body
Feb
5
asked What kind of errors arise when I fit ARMA(1,1) to data generated from ARMA(1,1)-GARCH(1,1) process?
Jan
21
comment George Soros models
does this result hold also for time series models like GARCH (for example) or is the result derived strictly from concept of linear regression model ?
Jan
18
accepted Problems with dealing with GARCH models and intra-day data
Nov
25
comment Switching from C++ to R - limitations/applications
try Julia julialang.org
Nov
18
comment How to estimate the following model?
I would draw attention, that in many situations, just "fitting" (not "estimating") model is needed. Fitting in the sense that we won't need to obtain variance of parameters and so their p-values. Minimization of MSE is enought and is straightforward even in the case of family of GARCH model (but computationally demanding and multiple restarts are needed to make sure), MSE is ok, as it is M-estimator. After fitting model we can check its predictive power and goodness of fitness to train date, to get first glimpse about its usefullness. Its like fast prototyping.
Nov
15
awarded  Yearling
Nov
12
asked Filtering out AR(1) effects before using stochastic volatility model
Apr
8
asked Are power contracts traded on any stock market?
Apr
2
awarded  Necromancer
Dec
13
awarded  Nice Question
Mar
5
comment Toy models of asset returns
en.wikipedia.org/wiki/Agent-Based_Computational_Economics , and suitable process which generates density with excesse kurtosis could be obtain by sum of a random number of i.i.d. Gaussians - sum of random number of trades each trade moves price iid Gaussian, number of trades is random and arise from interaction between agents - I've seen a few such papers; for general overview of agent models in quantitative finanse you could check this survey hal.archives-ouvertes.fr/docs/00/62/10/59/PDF/reviewII.pdf ps. zero intelligence models are easier = starting point
Feb
26
asked Why C is still in use especially in area of numerical optimization (instead of C++)?
Feb
16
comment Problems with dealing with GARCH models and intra-day data
I didn't use fGarch because this implementation of garch does not support external variables and I haven't been aware of existance of rugarch package
Feb
16
comment Problems with dealing with GARCH models and intra-day data
but AFRIMA which also have property of long memory didn't work - $d$ was statistically insignificant