2. 2
Frequency domain approach
Examines contributions of different
frequencies in explaining the variance.
Analysis based on the estimated spectral
density function.
Provides the information on the properties of
the time series data.
Applied to econometric problems*.
3. 3
Example
Monthly growth rate of IP , T = 513
peaks at k = 18, 44, 89, 128, 171, 210
cycle vk = k/T = 18/513, 44/513, .., 210/513
period Tk = 1/vk = 28.5, 12, … months
(28.5/12=2.3 yrs business cycle, 12, ..
seasonality,, )
frequency wk = 2vk = 2(18/513), ..
(per unit time in radian)
4. 4
Use of the spectral density function
S(wk) of X, where wk = 2vk
Total area under the curve from 0 to
= .5 Var(X)
(Symmetric from to 2)
We examine if low or high frequency dominates.
Examples (using “PEST” program)
unit root process (low)
white noise (horizontal line)
Stationary MA(1), AR(1) process (high)
5. 5
Background
Fourier transformation
Xt= {over k=0 to T/2} Xt(vk)
= [akcos(wkt) + bksin(wkt)]
where ak and bk are orthogonal Fourier coefficients.
• Xt(vj) and Xt(vk) are orthogonal.
Variance decomposition
Var(Xt) = Var(Xt(vk)) = k
2
… The variance is decomposed over different frequencies.
6. 6
Another form of (Discrete) Fourier
transformation
X(k) = T-1 Xt exp(-iwkt) = Xc(k) - iXs(k)
Inverse Fourier transformation
Xt = sum {over k=-(T/2) to (T/2)} X(k)exp(iwkt)
Periodogram
I(wk) = 2T[Xc(k)2 + Xs(k)2]
.. Not-consistent estimator for the spectral density
Background
7. 7
Spectral density function
Spectral density function
Sx(wk) = (1/2) {over j= -to}j exp(-iwj)
= (1/2)[0 + 2 {over j=0 to} j cos(wj)]
fx(wk) = Sx(wk)/ 0
.. Normalized spectral density
Inversion
0 = integral {from - to}Sx(wk) dw
8. 8
Smoothed spectral density
Sx(wk) = (1/2) {over j= -to}j exp(-iwj)
= (1/2)[0 + 2 {over k=1 toM} wn(k)j cos(wj)]
where wn(k) is a lag window (kernel)
M is a bandwidth.
Note: Automatic bandwidth by Andrews(1991)
Spectral density function
9. 9
Applications to Econometrics
Spectral density at frequency zero
Sx(0) = (1/2) {over j= -to} j
“longrun variance” = 2 Sx(0)
2 = 0 + 2 {over k=1 toM} wn(k)j
… captures “unknown” error structure
(non-parametric estimation)
11. 11
Hannan’s efficient estimator
yt = Xt’ +ut with unknown autocorrelation
Transform yt & Xt in frequency domain, then
do OLS on the transformed variables, say yt* & Xt*.
Transformation is based on the cross spectral
density of
yt & ut (also, Xt & ut), then inverse transformation
Applications to Econometrics
12. 12
Goodness-of-fit test
.. Testing for a white noise process (or any ARMA)
Based on the cumulative peridogram
Max difference follows Kolmogorov-Smirnov
statistics.
Applications to Econometrics