General notation |
|
:= |
‘is defined as’ |
x+, x− |
max{x, 0}, max{ − x, 0} (or min{x, 0} in Chapter 10) |
Sets and spaces |
|
, , , ![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__211D_fmt.jpg) |
positive integers, integers, rational numbers, real numbers |
+ |
positive real line |
d |
d-dimensional Euclidean space |
Dc |
complement of the set D ⊂ d |
[a,b) |
half-closed interval |
Matrices |
|
Id |
d-dimensional Identity matrix |
p,q( ) |
the set of p × q real matrices |
Processes |
|
iid |
independent and Identically distributed |
iid (0,1) |
iid centered with unit variance |
(Xt) or (Xt)t![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__2208_fmt.jpg) ![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__2124_fmt.jpg) |
discrete-time process |
( t) |
GARCH process |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__sigma-2t_fmt.jpg) |
conditional variance or volatility |
(χt) |
strong white noise with unit variance |
κχ |
kurtosis coefficient of ηt |
L or B |
lag operator |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexii001_fmt.jpg) |
sigma-field generated by the past of Xt |
Functions |
|
double-struck-1A (x) |
1 if x A, 0 otherwise |
[x] |
integer part of x |
γx,ρx |
autocovariance and autocorrelation functions of (Xt) |
x. x |
sample autocovariance and autocorrelation |
Probability |
|
(m, Σ) |
Gaussian law with mean m and covariance matrix Σ |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexii002_fmt.jpg) |
chi-square distribution with d degrees of freedom |
(α) |
quantile of order α of the distribution |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexiv001_fmt.jpg) |
convergence in distribution |
a.s. |
almost surely |
υn= op(un) |
υn/un → 0 in probability |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexiv002_fmt.jpg) |
a equals b up to the stochastic order op(l) |
Estimation |
|
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__F109_fmt.jpg) |
Fisher information matrix |
(κη − 1)J− 1 |
asymptotic variance of the QML |
θ0 |
true parameter value |
Θ |
parameter set |
θ |
element of the parameter set |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexiv005_fmt.jpg) |
estimators of θ0 |
= (θ) |
volatility built with the value θ |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexiv003_fmt.jpg) |
as but with initial values |
= t |
− 2 log (conditional variance of t) |
![image](http://images-20200215.ebookreading.net/1/2/2/9780470683910/9780470683910__garch-models__9780470683910__images__flast-iexiv006_fmt.jpg) |
approximation of t, built with initial values |
VARas Coyas |
asymptotic variance and covariance |
Some abbreviations |
|
ES |
expected shortfall |
FGLS |
feasible generalized least squares |
OLS |
ordinary least squares |
QML |
quasi-maximum likelihood |
RMSE |
root mean square error |
SACR |
sample autocorrelation |
SACV |
sample autocovariance |
SPAC |
sample partial autocorrelation |
VaR |
value at risk |