Scribd
Upload
84 views44 pages

Emery & Thomson (2005) :: Chapter 5: Time Series Analysis Methods

This document discusses time series analysis methods for detecting periodicities in data. It presents time series data on sea surface height, sea surface temperature, and zonal and meridional ocean currents from 2007-2012. It examines the fundamental periods and amplitude strengths for these time series. It also decomposes the sea surface temperature time series to analyze inter-annual, annual, semi-annual, and intra-seasonal periodicities. Finally, it discusses Fourier analysis and how it can be used to write any function as an infinite sum of trigonometric functions, allowing detection of periodic signals within time series data.

Uploaded by

Andry Purnama Putra
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
84 views44 pages

Emery & Thomson (2005) :: Chapter 5: Time Series Analysis Methods

This document discusses time series analysis methods for detecting periodicities in data. It presents time series data on sea surface height, sea surface temperature, and zonal and meridional ocean currents from 2007-2012. It examines the fundamental periods and amplitude strengths for these time series. It also decomposes the sea surface temperature time series to analyze inter-annual, annual, semi-annual, and intra-seasonal periodicities. Finally, it discusses Fourier analysis and how it can be used to write any function as an infinite sum of trigonometric functions, allowing detection of periodic signals within time series data.

Uploaded by

Andry Purnama Putra
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 44

Emery & Thomson (2005) :

Chapter 5: Time Series Analysis Methods


Basic Concepts (375-378)
Correlation Function (p. 378-384)
Fourier Analysis (p.384-388)

Fourier Series

. any function could be written as an infinite sum of the trigonometric


functions cosine and sine..

purpose of the lecture

detect and quantify periodicities in data

importance of periodicities

Time Series SSH & SST (2007-2012)

Annual

Fundamental period
Strength of amplitude ?

Time Series Zonal & Merid Current (2007-2012)

Fundamental period ?
Strength of amplitude ?

Decomposition of SST time series

Periodicity:
Inter-annual,
Annual,
Semi-annual,
Intra-seasonal

Temporal Periodicities
and their periods
astronomical

other natural

anthropogenic

rotation

ocean waves

electric power

daily

a few seconds

60 Hz

revolution

ocean currents

..

yearly

days months

..

sinusoidal oscillation
f(t) = C cos{ 2 (t-t0) / T }
cosine example
3

amplitude, C

d(t)

d(t)

-1

-2

-3

period, T
0

10

20

30

40

50
time, t

60

time, t

delay, t0

70

80

90

Periodicities
temporal

spatial

f(t) = C cos{ 2 t / T }

f(x) = C cos{ 2 x / }

amplitude, C

amplitude, C

period, T

wavelength,

frequency, f=1/T

angular frequency, =2 /T

f(t) = C cos(t)

wavenumber, k=2 /

f(x) = C cos(kx)

spatial periodicities
and their wavelengths
natural

anthropogenic

sand dunes

furrows plowed
in a field

hundreds of meters

few tens of cm

tree rings
a few millimeters

pairing sines and cosines


to avoid using time delays

derived using trigonometri identity

B
2
2
A =C cos (t

A=C cos(t0)
0)
2
2
B=C sin(t0)
B =C sin (t0)


A2+B2=C2 [cos2 (t0)+sin2 (t0)]
= C2

Fourier Series
linear model containing nothing but
sines and cosines

s are auxiliary
variables

As and Bs are
model parameters

two choices

values of frequencies?
total number of frequencies?

surprising fact about time series with


evenly sampled data

Nyquist frequency

values of frequencies?
evenly spaced, n = (n-1)
minimum frequency of zero
maximum frequency of fny

total number of frequencies? N/2+1


number of model parameters, M
= number of data, N

implies

Number of Frequencies
why N/2+1 and not N/2 ?

first and last sine are omitted from the


Fourier Series since they are
identically zero:

cos(N t)
col 1
cos(0t)

time, s

-2
0

col 2
cos( t)

-2
0

col 3
sin(t)

-2
0

col 4
cos(2 t)

-2
0

-2
0

col 5
sin(2 t)

0

col 32
-2
0

10

10

10

10

10

10

15

15

15

15

15

15

20

20

20

20

20

20

25

25

25

25

25

25

30

30

30

30

30

30

Nyquist Sampling Theorem

another way of stating it


when m=n+N
note evenly
sampled times

problem of aliasing
high frequencies
masquerading as low frequencies
solution:
pre-process data to remove high
frequencies
before digitizing it

Discrete Fourier Series

d = Gm

Least Squares Solution


mest = [GTG]-1 GTd
has substantial simplification
since it can be shown that

frequency and time setup


in MatLab
% N = number of data, presumed even
% Dt is time sampling interval
t = Dt*[0:N-1];
Df = 1 / (N * Dt );
Dw = 2 * pi / (N * Dt);
Nf = N/2+1;
Nw = N/2+1;
f = Df*[0:N/2];
w = Dw*[0:N/2];

Building G in MatLab
% set up G
G=zeros(N,M);
% zero frequency column
G(:,1)=1;
% interior M/2-1 columns
for i = [1:M/2-1]
j = 2*i;
k = j+1;
G(:,j)=cos(w(i+1).*t);
G(:,k)=sin(w(i+1).*t);
end
% nyquist column
G(:,M)=cos(w(Nw).*t);

solving for model parameters


in MatLab

gtgi = 2* ones(M,1)/N;
gtgi(1)=1/N;
gtgi(M)=1/N;
mest = gtgi .* (G'*d);

how to plot the model parameters?


As and B s

plot
against frequency

power spectral density

big at frequency when


when sine or cosine at the frequency
has a large coefficient

alternatively, plot
amplitude spectral density

Tambahan .
Transformasi Fourier

2. Fourier Transform
Transformasi Fourier adalah transformasi matematika dengan
banyak aplikasi dalam fisika dan teknik.
Mengubah fungsi matematika waktu f(t) menjadi fungsi baru F atau
dengan argumen frekuensi dg satuan siklus/detik (hertz).
Fungsi baru ini dikenal sebagai Transformasi Fourier atau
spektrum frekuensi dari fungsi f.
f disebut domain waktu, dan F adalah domain frekuensi.
Transformasi Fourier dinyatakan sebagai :
untuk setiap bilangan riil (zeta)
Variabel bebas x mewakili waktu (detik) dan variabel transformasi
merupakan frekuensi (hertz).

Visualisasi dari Transformasi Fourier

Transformasi Fourier
menghubungkan fungsi
domain waktu f (ditampilkan
dalam warna merah)
time

Visualisasi dari Transformasi Fourier

terhadap fungsi domain


frekuensi (ditampilkan dalam
warna biru)

Visualisasi dari Transformasi Fourier

Komponen-komponen
frekuensi tersebar di seluruh
spektrum frekuensi

Visualisasi dari Transformasi Fourier

Komponen-komponen
frekuensi tersebar di seluruh
spektrum frekuensi

Visualisasi dari Transformasi Fourier

Komponen-komponen
frekuensi tersebar di seluruh
spektrum frekuensi

Visualisasi dari Transformasi Fourier

Yang ditampilkan sebagai


puncak dalam domain
frekuensi

frequency

Visualisasi dari Transformasi Fourier

Fungsi f dalam domain


waktu

Transformasi Fourier

time

Hasil transformasi Fourier


dalam domain frekuensi
frequency

Time-series arus zonal


27 day
14 day

Time-series arus zonal


27 day

14 day

Wattimena et al. (2014)

You might also like