爱气象,爱气象家园! 

气象家园

 找回密码
 立即注册

QQ登录

只需一步,快速开始

新浪微博登陆

只需一步, 快速开始

搜索
查看: 20350|回复: 44

奇异谱分析软件工具包(SSA-MTM Toolkit)介绍

  [复制链接]

新浪微博达人勋

发表于 2013-12-22 12:15:51 | 显示全部楼层 |阅读模式

登录后查看更多精彩内容~

您需要 登录 才可以下载或查看,没有帐号?立即注册 新浪微博登陆

x
本帖最后由 hillside 于 2013-12-22 12:28 编辑

        该工具包来自加州大学洛杉矶分校大气海洋科学系网页,可填写EMail等信息申请下载。
                                                              http://www.atmos.ucla.edu/tcd/ssa/
        User's Guide  Latest version (version 4.4)
  • Frequently asked questions
  • Bibliographic references to SSA and MTMWho we areOther info
    What it is / What it does:The Singular Spectrum Analysis - MultiTaper Method (SSA-MTM) Toolkit is a software program to analyze short, noisy time series, such as the one below, as well as multivariate data.

                                   
    登录/注册后可看大图
    This is the so-called Southern Oscillation Index (SOI). SOI is a climatic index connected with the recurring El NiÑo conditions in the tropical Pacific; it is essentially the normalized monthly mean difference in sea-level pressure between Darwin, Australia and Tahiti (Rasmusson et al., 1990).
    Using a convinient graphical user interface you can perform singular-spectrum analysis on input data.


                                   
    登录/注册后可看大图

    SSA reconstructions of selected components and tests for the presence of trend and oscillatory components are provided. Ad hoc and Monte Carlo error bars for the SSA eigenspectra are also included. All results are displayed graphically. The next figure shows SSA eigenspectrum for the SOI time series:


                                   
    登录/注册后可看大图
    In addition, the Toolkit includes three kinds of power-spectrum estimation. These are the traditional Blackman-Tukey windowed correlogram, multi-taper method(MTM), and maximum-entropy method(MEM). You can apply these tools at any point in the analysis to a raw time series, or to SSA reconstructions. Outputs include power spectra and significance tests for correlogram and MTM.
    The next figure shows MTM spectrum for the same data. The black peaks are components of the spectrum associated with a periodic signal, and the significance levels relative to the estimated noise background are shown.


                                   
    登录/注册后可看大图

    Release Notes:
    • September 17, 2013: Minor bugs fixed.
    • January 28, 2013: Major upgrade of multivariate analysis: Varimax Rotationfor M-SSA . Available on Linux and Mac OS builds.
      Also, beta-version (with limited features, currently mostly M-SSA) of command-line Toolkit utilities are available for Linux and Mac OS;choose "CMD for Linux" or "CMD for Mac", respectively, in download options.
    • January 4, 2013: Updated 64-bit Linux and 64-bit Mac OS.
    • February 4, 2008: New 64-bit bnaries for Linux and Mac OS 10.5 (Intel); 32-bit binaries for Linux and Mac OS X (Intel) recompiled with Intel compiler .
    • April 12, 2007: Toolkit 4.4 (includes SSA gap-filling)!
    • May 16, 2006: Universal Binary of Toolkit 4.3 for Mac OS X!
    • November 11, 2005: Toolkit 4.3
      • Bug, leading to crashes in MTM Reconstruction, has been fixed.
      • Bug, leading to getting zero vector for MTM Reconstruction ofvery low-frequencies, has been fixed.
    • September 26, 2003: Toolkit 4.2 (includes MSSA)!
    • September 4, 2002: Toolkit ported to Mac OS 10.2 (Jaguar)!
    • May 13, 2002: Minor bug is fixed in MTM when using non-deafult time step.
    • March 22, 2002: Bug fixed in computing MTM with varying frequency limits.
    • January 28, 2002: Mac OS X version updated for XFree86 4.2 release!
    • January 23, 2002: Mac OS X version updated!(Much faster computations, check it out)
    • April 11, 2001: Version 4.1 ported to Mac OS X!
    • January 9, 2001: Version 4.1 ported to HP-UX!
    • September 28, 2000: Version 4.1


        • Graphics support for IDL (version 5.0 and later) and Grace plotting tools is added. See this guide section for more info.
        • Computation speed for Monte-Carlo SSA has been optimized by fixing resolution at 0.001/dt (in the Nyquist interval from 0 to 0.5/dt) for finding dominant frequencies for EOFs.
    • June 19, 2000: Version 4.0 update! MTM Confidence levels for the "white noise" null-hypothesis are plotted correctly now.
    • April 5, 2000: Toolkit 4.0!

    • Toolkit 4.0 is provided with a new graphical user interface (GUI) which is no longer based on TCl/Tk. 4.0 GUI is based on Motif, which is the industry standard on Unix computer systems. New GUI makes a Tollkit more flexible and easy to use.
    • Dynamical memory management is implemented to remove the restrictions on the time series length, existing in the old versions of the Toolkit. The results and data are managed using matrices and vectors with the names supplied by the user in GUI.
    • Linear algebra routines are now based on a latest release of LAPACK package (Version 3.0, June 1999). A considerable speed-up in Toolkit performance has been achieved.
    • New plotting routines are available which allow cross-examination of the Toolkit results.
    • Several minor bugs have been fixed including normalization for MTM Spectrum and choice of parameters for BT Correlogram.
  • Availability:Prebuilt binary executables for the Toolkit are available for download. The Toolkit has been ported and actively maintained for Linux and Mac OS X (under X11, runs natively on both PowerPC and Intel based Mac systems). Also, legacy builds for SUN, DEC and SGI are available (not all latest features may be included).Who we are:The SSA-MTM Toolkit is a product of the SSA-MTM Group (so far: Myles Allen, [url=http://www.atmos.ucla.edu/tcd/ssa/%20http://meteora.ucsd.edu/~dettinge]Mike Dettinger[/url], Kayo Ide, Dmitri Kondrashov, Michael Ghil, Mike Mann,Andrew W. Robertson, Amira Saunders, Ferenc Varadi, Yudong Tian, and Pascal Yiou) at UCLA (mostly). Andreas Groth helped to include VARIMAX rotation in M-SSA. Command-line version of the Toolkit is developed by Bruno Deremble and Dmitri Kondrashov.
    You can direct comments on installation and performance, as well as suggestions for future versions, to ssahelp@atmos.ucla.edu.

    Other information:
    • For the application of the Toolkit in the life and biomedical sciences, please see:

      A. Brawanski, R. Faltermeier, R. D. Rothoerl, C. Woertgen, 2002:
      Comparison of near-infrared spectroscopy and tissue PO2 time series in
      patients after severe head injury and aneurysmal subarachnoid hemorrhage.
      J. Cerebr. Blood F. Met., 22 (5): 605--611.
    Colebrook, J.M., 1978:
    Continuous plankton records - zooplankton and environment,
    northeast Atlantic and North Sea, 1948-1975,
    Oceanol. Acta ,1, 9-23.
    M. Grigorov, 2006: Global dynamics of biological systems from
    time-resolved omics experiments, Bioinformatics, 22 (12), 1424--1430.

    Mineva A, Popivanov D, 1996: Method of singletrial readiness potential
    identification, based on singular spectrum analysis.
    J. Neurosci. Methods, 68, 91-99.

    Rodo X, Pascual M, Fuchs G, Faruque ASG, 2002:
    ENSO and cholera: A nonstationary link related to climate change?
    Proc. Nat. Acad. Sci. USA, 99 (20), 12901-12906.

  • Grant Support:
1. ONR-N00014-12-1-0911, FY2012 Multi-University Research Initiative (MURI) Topic #16: Extended-Range Environmental Prediction Using Low-Dimensional Dynamic Modes, Office of Naval Research, 2012--2015.
2. NSF 1049253, Collaborative Research, Type 1, L02170206: Climate Sensitivity, Stochastic Models and GCM-EaSM Optimization, U.S. National Science Foundation (DMS + MPS Divisions), 2011--2014.
3. DOE DE-SC0006694, Decadal Prediction and Stochastic Simulation of Hydroclimate over Monsoonal Asia , U.S. Department of Energy, 2011--2014.

General Questions1) What is SSA-MTM Toolkit?2) How do I reference SSA-MTM Toolkit in publication?3)Can the Toolkit handle unevenly sampled time-series data (a varying time step)?4) Do I need to normalize and/or center the time series?5) How can I tell if what I'm seeing is a genuine signal or an artefact of the noise?6) I would like to know if we can use the SSA Toolkit for forecasting?7) Is there a Toolkit version that allows to analyse multi-channel data?Installation & Running8)I have downloaded file Spectra*.exe, what should I do with it?9)I am getting weird messages like this when starting Spectra:"...ld.so.1: Spectra: fatal:", "libF77.so.4: open failed:...", "libg2c.so.0: cannot load....."?10)Show me the Toolkit release notes?11)What are the system requirements for running the Toolkit?12)What visualization software is supported by the Toolkit?13) Is there an easy way to adjust the default size of the xmgr x-window?14)The system reports `Cannot find xmgr to execute'.15)How can I save results to a file?16) We can load data and do computations with it; however nothing is happening when we press Plot button. The software for plotting, Grace for example, has been installed17)I am getting message "error during reading file" when trying to read input data.


General Questions1) What is SSA-MTM Toolkit?
The Singular Spectrum Analysis - Multitaper Method (SSA-MTM) Toolkit consists of a set of programs that perform detailed spectral analyses and decompositions on a univariate input time series. The toolkit contains procedures for:






    (a) estimating the spectrum of a time series,






    (b) decomposing the time series into trends, oscillatory components, and noise, and






    (c) reconstructing the contributions of selected components of the time series.
Four methods of spectral-analysis (task a) are provided: Blackman-Tukey correlogram estimation (BT), the Maximum-Entropy Method (MEM), the Multi-Taper Method (MTM), and Singular-spectrum Analysis (SSA). SSA and MTM can then be used for separation of trends, near-periodic and other significant components, and noise (tasks b and c). Both SSA and MTM incorporate sophisticated significance tests against a variety of noise null-hypotheses.
2) How do I reference SSA-MTM Toolkit in publication?

If you feel that your research has benefitted from the use of the SSA-MTM Toolkit, you can repay us by citing our articles:


Ghil M., R. M. Allen, M. D. Dettinger, K. Ide, D. Kondrashov, M. E. Mann, A. Robertson, A. Saunders, Y. Tian, F. Varadi, and P. Yiou, 2002: "Advanced spectral methods for climatic time series," Rev. Geophys.,40(1), pp. 3.1-3.41, 10.1029/2000RG000092.
Dettinger, M.D., Ghil, M., Strong, C.M., Weibel, W., and Yiou, P., 1995: Software expedites singular-spectrum analysis of noisy time series, Eos, Trans. American Geophysical Union, v. 76(2), p. 12, 14, 21.We also appreciate your references to the original articles that motivated and made the Toolkit possible:
  • Allen, M.R., and Smith, L.A., 1996: Monte Carlo SSA: detecting oscillations in the presence of coloured noise. J. Climate, 9, 3373.
  • Mann, M.E. and Lees, J.M., 1996: Robust estimation of background noise and signal detection in climatic time series, Clim. Change, 33, 409-445.
  • Vautard, R., Yiou, P., and Ghil, M., 1992: Singular-spectrum analysis: A toolkit for short, noisy chaotic signals, Physica D, 58, 95-126.
3) Can the SSA Toolkit handle unevenly sampled time-series data (a varying time step)? Can the time information be used in the Toolkit in order to do the analysis, or does the data need to be interpolated to an even time-step before using the Toolkit?Gap-filling by SSA provided by the Toolkit can handle unevenly sampled data or even occasional missing data.
4) Do I need to normalize and/or center the time series?
No, this will be done inside the tools.

5) How can I tell if what I'm seeing is a genuine signal or an artefact of the noise?
There is no stock answer for this question. The basic philosophy of the Toolkit is that only the simultaneous and flexible application of more than one spectral estimation method can provide truely reliable information on a given time series, when the signal-to-noise ratio is low.
6) I would like to know if we can use the SSA Toolkit for forecasting.
SSA CAN be used for prediction, but our Toolkit doesn't perform that function yet.
To learn more about using SSA for prediction, you should see:
Keppenne, C.L., and Ghil, M., 1992, Adaptive filtering and prediction of the Southern Oscillation Index: J. Geophysical Research (atmospheres) 97, 20449-20454.
Jiang, N., Ghil, M., and Neelin, D., 1995, Quasi-quadrennial and quasi-biennial variability in the equatorial Pacific: Clim. Dyn. 12, 101-112.
7) Is there a Toolkit version that allows to analyse multi-channel data?
It is included in version 4.2 and higher

Installation and Running8)I have downloaded file Spectra*.exe, what should I do with it?The extension *.exe is needed to save the file using a web browser. You may rename the file directly in the dialog box or after you saved it, i.e. toSpectra. Now you need to give execute permission to the file:
chmod +x Spectra
Then you just type
./Spectra &
9)I am getting weird messages like this when starting Spectra:"...ld.so.1: Spectra: fatal:", "libF77.so.4: open failed:...", "libg2c.so.0: cannot load....."?
You are trying to run a dynamic version, and a particular compiler library is not installed. Try a static version, if it is available. In some cases you may need to change the environment variable LD_LIBRARY_PATH which tells where to look for the dynamic libraries.
10)Show me the Toolkit release notes?
You may see it here.




11)What are the system requirements for running the the Toolkit?
Currently the Toolkit has been ported to most Unix based systems, including SUN, IBM/AIX, DEC, SGI & Mac OS X. Also, Linux version for PC is avialable. You may download prebuilt binary executables (1~5Mb size) for supported systems.
12)What visualization software is supported by the Toolkit?
Toolkit supports the public domain ACE/gr (known also as Xmgr), descendant of ACE/gr called Grace, and commercial IDL (versions 5.0 and later) graphics software.
13) Is there an easy way to adjust the default size of the xmgr x-window?
The easiest way is probably to put a resource in ~/.Xdefaults. For example:
xmgr*geometry: 500x20014)The system reports `Cannot find xmgr to execute'.
The Toolkit is trying to plot something, but the plotter is not in your path. If you know that xmgr exists on your system, then make sure it is in your path.
15)How can I save results to a file?
You can write matrices/vectors into a file using 'Write Matrix' and 'Write Vector' options in 'File/Data' menu on the main panel. You just need to specify the name of the matrix/vector and the name of the file. Also, when you have xmgr plot on your screen, you may go to the Data/Status option and see what is/are the work file name(s) in the plot (F5 key does that too). Then you can copy it/them in a unix shell to a new file(s) you want to keep. (remember that all of the files with the extension *.tmp will be deleted when exiting Spectra).
16)We can load data and do computations with it; however nothing is happening when we press Plot button.
The software for plotting, Grace for example, has been already installed.

You need to add Grace to your path. You can do it in your working shell's configuration file, and the exact command depends on the particular shell used. For tcsh shell you have to add new line in .tcshrc file like below (the exact path depends, of course, on where you installed Grace):
set path= ("/usr/grace-5.1.12/grace/bin" $path)
then do from a command line
source .tcshrc
and start Toolkit from that shell.
17)I am getting message "error during reading file" when trying to read input data.Make sure that file is plain text in ASCII format, doesn't have blank spaces amd contains only numerical data. If it is matrix, it should have the same number of entries in all columns. If you are using Mac OS X, there should be no blank spaces in the path to the file.


Copyright © SSA-MTM group, (mostly) UCLA.

气象家园相关帖子:
[程序设计] SSA-MTM软件下载后,为什么安装不了??
[源代码] SSA源代码(fortran)
[程序设计] SSA奇异普分析

奇异谱分析软件工具包手册(users.guide3).tar.gz

237.04 KB, 下载次数: 124, 下载积分: 金钱 -5

点评

此软件2016年已经更新,可在windows下利用Cygwin运行~ http://research.atmos.ucla.edu/tcd//ssa/Windows_Cygwin.pdf  发表于 2017-3-21 11:15
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2013-12-22 13:20:41 | 显示全部楼层
这个很好!感谢!
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2014-3-8 11:56:30 | 显示全部楼层
好东西





















111
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2014-4-21 22:00:06 | 显示全部楼层
楼主,我找了半天都没有找到软件下载的地方,还有填写邮件申请的地方,能否指点一下,谢谢哦,都看说明看疯了
密码修改失败请联系微信:mofangbao

新浪微博达人勋

 楼主| 发表于 2014-4-21 22:06:24 | 显示全部楼层
本帖最后由 hillside 于 2014-4-21 22:09 编辑
lss_lucky 发表于 2014-4-21 22:00
楼主,我找了半天都没有找到软件下载的地方,还有填写邮件申请的地方,能否指点一下,谢谢哦,都看说明看疯 ...
我没有用过,只是介绍。以下是我浏览之下的认识(直接点击下面第3条后即可进行操作):
1.Availability:
Prebuilt binary executables for the Toolkit are available for download.

2.http://www.atmos.ucla.edu/tcd/ssa/form.html

3.点击页面出现的“downloadtoolkit”
4.提示填写EMAIL






密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2014-4-21 22:53:09 | 显示全部楼层
hillside 发表于 2014-4-21 22:06
我没有用过,只是介绍。以下是我浏览之下的认识(直接点击下面第3条后即可进行操作):
1.Availability:
...

谢谢啦,终于找到了,现在为版本问题纠结中,不知道要不要选liux
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2014-4-21 22:53:19 | 显示全部楼层
hillside 发表于 2014-4-21 22:06
我没有用过,只是介绍。以下是我浏览之下的认识(直接点击下面第3条后即可进行操作):
1.Availability:
...

谢谢啦,终于找到了,现在为版本问题纠结中,不知道要不要选liux
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2015-7-24 19:20:00 | 显示全部楼层
,奇异谱分析,有机会读读
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2015-7-31 15:05:58 | 显示全部楼层
这个很好!感谢!{:eb502:}
密码修改失败请联系微信:mofangbao

新浪微博达人勋

发表于 2015-8-28 10:18:37 | 显示全部楼层
学习一下怎么用
密码修改失败请联系微信:mofangbao
您需要登录后才可以回帖 登录 | 立即注册 新浪微博登陆

本版积分规则

Copyright ©2011-2014 bbs.06climate.com All Rights Reserved.  Powered by Discuz! (京ICP-10201084)

本站信息均由会员发表,不代表气象家园立场,禁止在本站发表与国家法律相抵触言论

快速回复 返回顶部 返回列表