Doesn’t look like ts-groups takes picture :-( I put the picture here
http://home.bigblue.net.au/~paulho/public_html/public_html/ARIMA.png
I want to work out the variance of the predicted price and hence the confidence
interval. I cant find any literature but derived this myself as shown below:-
The C++ code is as show below also. Am I correct?
//
***************************************************************************************************************
// * This inline function calculates the recurive noise variance during forward
prediction
// * e'(n+1) = e (n+1) + Sum (e'(n - k) * a(k)
// * Input : variance and ema are arrays of at least size extraF, a is the
alpha in ema smoothing
// * Input : extraF is the prediction length, order is order of coefficient,
ar is the coefficient array[degree+1]
// * Output : ema average of the noise variance at the last prediction bar
// *
//
***************************************************************************************************************
inline double forErr( int extraF, int order, double noise, double *ar, float a,
double *variance, double *ema)
{
variance[0] = ema[0] = noise;
for(int i = 1; i < extraF; i++)
{
variance[i] = noise;
for(int j = 1; j < order + 1; j++ )
{
double error = i - j < 0 ? 0.0f :
variance[i - j];
variance[i] += error * SQR(ar[j]);
}
ema[i] = a * variance[i] + (1 - a) * ema[i -1];
}
double result = ema[extraF - 1];
return result;
}
_____
From: [email protected] [mailto:[EMAIL PROTECTED] On Behalf Of Paul
Ho
Sent: Wednesday, 29 November 2006 11:00 PM
To: [email protected]
Subject: RE: [amibroker-ts] confidence level asymptoe of Predicted AR
Pictures gone missing;(
The errors of each Yn such that Y’n+1 = Yn+1 + en+1 Y’ is the approximation of
Y, and Y’n+1 = ∑ Yn-k . a k
The errors associated with Y”n+1 which is the second approximation using Y’k is
e’n+1 = en+1 + ∑ e’n-k . a k
am I correct?
_____
From: amibroker-ts@ <mailto:amibroker-ts%40yahoogroups.com> yahoogroups.com
[mailto:amibroker-ts@ <mailto:amibroker-ts%40yahoogroups.com> yahoogroups.com]
On Behalf Of Paul Ho
Sent: Wednesday, 29 November 2006 12:17 AM
To: amibroker-ts@ <mailto:amibroker-ts%40yahoogroups.com> yahoogroups.com
Subject: [amibroker-ts] confidence level asymptoe of Predicted AR
Hi
I want to get the confidence level asymptoes of the predicted AR similarly
to that shown in the picture
Can anybody help?
[Non-text portions of this message have been removed]
[Non-text portions of this message have been removed]
[Non-text portions of this message have been removed]