Visualising forecasting algorithm performance using time series instance spaces

It is common practice to evaluate the strength of forecasting methods using collections of well-studied time series datasets, such as the M3 data. The question is, though, how diverse and challenging are these time series, and do they enable us to study the unique strengths and weaknesses of differe...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International journal of forecasting Ročník 33; číslo 2; s. 345 - 358
Hlavní autoři: Kang, Yanfei, Hyndman, Rob J., Smith-Miles, Kate
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.04.2017
Témata:
ISSN:0169-2070, 1872-8200
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract It is common practice to evaluate the strength of forecasting methods using collections of well-studied time series datasets, such as the M3 data. The question is, though, how diverse and challenging are these time series, and do they enable us to study the unique strengths and weaknesses of different forecasting methods? This paper proposes a visualisation method for collections of time series that enables a time series to be represented as a point in a two-dimensional instance space. The effectiveness of different forecasting methods across this space is easy to visualise, and the diversity of the time series in an existing collection can be assessed. Noting that the diversity of the M3 dataset has been questioned, this paper also proposes a method for generating new time series with controllable characteristics in order to fill in and spread out the instance space, making our generalisations of forecasting method performances as robust as possible.
AbstractList It is common practice to evaluate the strength of forecasting methods using collections of well-studied time series datasets, such as the M3 data. The question is, though, how diverse and challenging are these time series, and do they enable us to study the unique strengths and weaknesses of different forecasting methods? This paper proposes a visualisation method for collections of time series that enables a time series to be represented as a point in a two-dimensional instance space. The effectiveness of different forecasting methods across this space is easy to visualise, and the diversity of the time series in an existing collection can be assessed. Noting that the diversity of the M3 dataset has been questioned, this paper also proposes a method for generating new time series with controllable characteristics in order to fill in and spread out the instance space, making our generalisations of forecasting method performances as robust as possible.
Author Kang, Yanfei
Hyndman, Rob J.
Smith-Miles, Kate
Author_xml – sequence: 1
  givenname: Yanfei
  surname: Kang
  fullname: Kang, Yanfei
  email: yanfei.kang@outlook.com
  organization: School of Economics and Management, Beihang University, Beijing, 100191, China
– sequence: 2
  givenname: Rob J.
  surname: Hyndman
  fullname: Hyndman, Rob J.
  email: Rob.Hyndman@monash.edu
  organization: Department of Econometrics and Business Statistics, Monash University, Clayton VIC 3800, Australia
– sequence: 3
  givenname: Kate
  surname: Smith-Miles
  fullname: Smith-Miles, Kate
  email: Kate.Smith-Miles@monash.edu
  organization: School of Mathematical Sciences, Monash University, Clayton VIC 3800, Australia
BookMark eNqNkNtKAzEQhoNUsK2-w77ArpPNHm8ELZ5A6I16G7LZSZ1lDyVJBd_etFUEb_RqhvlnPphvwWbjNCJjEYeEAy8uu4Q6M1nUyvkkDZME6gQgO2FzXpVpXKUAMzYPQR2nUMIZWzjXAUBecj5n61dyO9WTo3ETfXP2veo3kyX_NkRbtCEY1Kgx2h32PA0YObSELqLR-UPktkqjO2enRvUOL77qkr3c3T6vHuKn9f3j6vop1qISPm4q0SIYg9BWaQUpcpNBm4mqybNatY2um9wgFnldirwQjQBu2jwrNWRNUfNCLFl15Go7OWfRyK2lQdkPyUHuxchO_oiRezESahnEhNOrX6eavPI0jd4q6v8DuDkCMDz4Tmil04TBQUth28t2or8hnzUIit4
CitedBy_id crossref_primary_10_1016_j_ijforecast_2020_11_009
crossref_primary_10_2139_ssrn_5122312
crossref_primary_10_1016_j_ijforecast_2025_01_001
crossref_primary_10_1007_s10994_022_06205_9
crossref_primary_10_1155_2022_6206037
crossref_primary_10_1002_widm_1404
crossref_primary_10_1111_anzs_12441
crossref_primary_10_3390_w15040634
crossref_primary_10_1016_j_ijforecast_2022_11_005
crossref_primary_10_1080_10618600_2019_1617160
crossref_primary_10_1016_j_ijforecast_2021_11_001
crossref_primary_10_1080_10618600_2020_1807997
crossref_primary_10_1016_j_patcog_2021_108441
crossref_primary_10_1016_j_jhydrol_2023_129160
crossref_primary_10_3390_a16040206
crossref_primary_10_1016_j_scitotenv_2020_144612
crossref_primary_10_1080_00207543_2022_2153941
crossref_primary_10_1016_j_compag_2024_109124
crossref_primary_10_1038_s41597_020_0553_0
crossref_primary_10_1145_3436893
crossref_primary_10_1016_j_ijforecast_2022_07_010
crossref_primary_10_1287_ijds_2021_0003
crossref_primary_10_1145_3447687
crossref_primary_10_1016_j_ijforecast_2021_07_006
crossref_primary_10_3390_a14030095
crossref_primary_10_3389_frwa_2022_961954
crossref_primary_10_1016_j_ijforecast_2021_07_002
crossref_primary_10_1016_j_ijforecast_2020_07_007
crossref_primary_10_1162_evco_a_00262
crossref_primary_10_1080_01605682_2022_2118629
crossref_primary_10_1016_j_ijforecast_2022_06_004
crossref_primary_10_1186_s40645_023_00574_y
crossref_primary_10_1145_3572895
crossref_primary_10_1016_j_ejor_2020_09_046
crossref_primary_10_1016_j_apenergy_2021_117242
crossref_primary_10_1016_j_eswa_2023_122461
crossref_primary_10_1080_26941899_2025_2517006
crossref_primary_10_1080_19427867_2024_2310831
crossref_primary_10_3390_rs14153821
crossref_primary_10_1007_s10489_024_05715_4
crossref_primary_10_5194_amt_13_1019_2020
crossref_primary_10_1016_j_ijforecast_2019_02_011
crossref_primary_10_1007_s10994_021_05959_y
crossref_primary_10_1016_j_ijforecast_2019_02_012
crossref_primary_10_1109_ACCESS_2020_2971591
crossref_primary_10_1016_j_ijforecast_2019_02_013
crossref_primary_10_3390_en17164163
crossref_primary_10_1016_j_ijforecast_2019_02_014
crossref_primary_10_1007_s10472_020_09695_2
crossref_primary_10_1016_j_ijforecast_2018_06_001
crossref_primary_10_1080_01605682_2021_1880297
crossref_primary_10_1007_s10994_021_06126_z
crossref_primary_10_1016_j_eswa_2020_113680
crossref_primary_10_1016_j_ijforecast_2019_06_001
crossref_primary_10_3390_e22010089
crossref_primary_10_1016_j_apenergy_2024_125176
crossref_primary_10_1016_j_gsf_2022_101349
crossref_primary_10_1111_2041_210X_13915
crossref_primary_10_3390_w14101657
crossref_primary_10_1145_3519313
crossref_primary_10_1002_for_2963
crossref_primary_10_1145_3699596
crossref_primary_10_3390_bdcc7020100
crossref_primary_10_1109_TSE_2022_3228334
crossref_primary_10_1007_s11047_021_09847_1
crossref_primary_10_1016_j_ijforecast_2020_02_005
crossref_primary_10_1080_00207543_2021_2022800
crossref_primary_10_1007_s10618_022_00826_3
crossref_primary_10_1016_j_neucom_2025_131183
crossref_primary_10_1080_00207543_2023_2227903
crossref_primary_10_1016_j_jbusres_2020_10_051
crossref_primary_10_1016_j_apm_2025_115965
crossref_primary_10_3389_feart_2024_1511785
crossref_primary_10_1016_j_knosys_2020_106467
crossref_primary_10_1016_j_cor_2024_106747
crossref_primary_10_1007_s13222_018_00304_5
crossref_primary_10_1080_00207543_2022_2063086
crossref_primary_10_1016_j_rser_2021_110735
crossref_primary_10_3390_rs15204912
crossref_primary_10_1109_TPAMI_2020_3038760
crossref_primary_10_1145_3640335
crossref_primary_10_1287_mnsc_2022_4485
crossref_primary_10_3727_108354220X16002732379690
crossref_primary_10_1016_j_ijforecast_2019_03_015
crossref_primary_10_1088_2632_2153_ad63f3
crossref_primary_10_1016_j_ijforecast_2019_03_017
crossref_primary_10_1177_10946705241232169
crossref_primary_10_1007_s10618_021_00745_9
crossref_primary_10_1016_j_ijforecast_2018_10_004
crossref_primary_10_1007_s00521_021_06244_8
crossref_primary_10_1016_j_cor_2020_105184
crossref_primary_10_1016_j_ijepes_2022_108092
crossref_primary_10_3390_app13127065
crossref_primary_10_1007_s10479_024_05882_0
crossref_primary_10_1007_s10618_019_00661_z
crossref_primary_10_1016_j_iot_2021_100432
crossref_primary_10_1016_j_segan_2023_101171
crossref_primary_10_1016_j_softx_2025_102246
crossref_primary_10_1016_j_eswa_2025_126798
crossref_primary_10_1007_s11063_022_11113_z
crossref_primary_10_1371_journal_pone_0194889
crossref_primary_10_1016_j_patcog_2022_109014
crossref_primary_10_3390_forecast3010010
crossref_primary_10_1029_2024EA003966
crossref_primary_10_1016_j_patcog_2022_109132
crossref_primary_10_1007_s10586_024_04909_2
crossref_primary_10_1016_j_eswa_2022_117584
crossref_primary_10_1016_j_ijforecast_2018_12_007
crossref_primary_10_1002_sam_11461
crossref_primary_10_1080_10407413_2022_2055473
crossref_primary_10_1016_j_ijforecast_2019_05_002
Cites_doi 10.1016/S0304-4076(01)00125-7
10.1016/j.ins.2013.02.030
10.1016/S0169-2070(00)00057-1
10.1109/4235.585893
10.1111/j.2517-6161.1964.tb00553.x
10.1103/PhysRevE.90.052910
10.1016/j.cor.2013.11.015
10.1175/JAS-D-13-0182.1
10.1103/PhysRevLett.88.174102
10.1016/S0169-2070(01)00110-8
10.1016/j.ecolind.2011.07.008
10.18637/jss.v053.i04
10.1103/PhysRevE.87.022911
10.1016/S0169-2070(00)00066-2
10.1016/j.ijforecast.2006.03.001
10.1016/j.ejor.2014.02.036
10.1098/rsif.2013.0048
10.1007/s10618-005-0039-x
10.1109/TKDE.2014.2316504
10.1002/qj.2501
10.1016/j.cor.2015.04.022
10.1162/neco.1996.8.7.1341
ContentType Journal Article
Copyright 2016 International Institute of Forecasters
Copyright_xml – notice: 2016 International Institute of Forecasters
DBID AAYXX
CITATION
DOI 10.1016/j.ijforecast.2016.09.004
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Social Sciences (General)
EISSN 1872-8200
EndPage 358
ExternalDocumentID 10_1016_j_ijforecast_2016_09_004
S0169207016301030
GroupedDBID --K
--M
-~X
.L6
.~1
0R~
13V
1B1
1OL
1RT
1~.
1~5
29J
3R3
4.4
457
4G.
5GY
5VS
63O
7-5
71M
85S
8P~
96U
9JO
AAAKF
AAAKG
AACTN
AAEDT
AAEDW
AAFFL
AAIAV
AAIKJ
AAKOC
AALRI
AAOAW
AAPFB
AAQFI
AAQXK
AARIN
AAXUO
ABEHJ
ABJNI
ABKBG
ABLJU
ABMAC
ABMVD
ABTAH
ABUCO
ABXDB
ABYKQ
ACBMB
ACDAQ
ACGFO
ACGFS
ACHQT
ACHRH
ACNTT
ACRLP
ACROA
ADBBV
ADEZE
ADFHU
ADMUD
AEBSH
AEKER
AEYQN
AFAZI
AFFNX
AFKWA
AFODL
AFTJW
AGHFR
AGJBL
AGTHC
AGUBO
AGUMN
AGYEJ
AHHHB
AI.
AIEXJ
AIIAU
AIKHN
AITUG
AJBFU
AJOXV
AJWLA
ALEQD
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
APLSM
ASPBG
AVWKF
AXJTR
AXLSJ
AZFZN
BEHZQ
BEZPJ
BGSCR
BKOJK
BKOMP
BLXMC
BNSAS
BNTGB
BPUDD
BULVW
BZJEE
CS3
DU5
EBS
EFJIC
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-2
G-Q
GBLVA
HAMUX
HLX
HVGLF
HZ~
IHE
IXIXF
J1W
KOM
LG8
LPU
LXL
LXN
LY1
M41
MO0
MS~
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
PQQKQ
Q38
R2-
RIG
ROL
RPZ
SBM
SDF
SDG
SDP
SDS
SEB
SES
SEW
SPCBC
SSB
SSD
SSF
SSL
SSZ
T5K
TN5
U5U
VH1
WUQ
XPP
XYO
YK3
ZMT
ZRQ
ZY4
~G-
9DU
AATTM
AAXKI
AAYWO
AAYXX
ABWVN
ACLOT
ACRPL
ACVFH
ADCNI
ADMHG
ADNMO
AEIPS
AEUPX
AFJKZ
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKRWK
AKYEP
ANKPU
APXCP
CITATION
EFKBS
~HD
ID FETCH-LOGICAL-c383t-b83de0ffe0d82802e1f40d438b549adbc9b5fee65973563b301fd547c04b69163
ISICitedReferencesCount 119
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000399512200002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0169-2070
IngestDate Sat Nov 29 03:06:45 EST 2025
Tue Nov 18 22:34:20 EST 2025
Fri Feb 23 02:37:28 EST 2024
IsPeerReviewed true
IsScholarly true
Issue 2
Keywords M3-Competition
Forecasting algorithm comparison
Time series visualisation
Time series generation
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c383t-b83de0ffe0d82802e1f40d438b549adbc9b5fee65973563b301fd547c04b69163
PageCount 14
ParticipantIDs crossref_primary_10_1016_j_ijforecast_2016_09_004
crossref_citationtrail_10_1016_j_ijforecast_2016_09_004
elsevier_sciencedirect_doi_10_1016_j_ijforecast_2016_09_004
PublicationCentury 2000
PublicationDate 2017-04-01
PublicationDateYYYYMMDD 2017-04-01
PublicationDate_xml – month: 04
  year: 2017
  text: 2017-04-01
  day: 01
PublicationDecade 2010
PublicationTitle International journal of forecasting
PublicationYear 2017
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Hyndman, Athanasopoulos (br000080) 2014
Hyndman, Koehler, Snyder, Grose (br000095) 2002; 18
Assimakopoulos, Nikolopoulos (br000005) 2000; 16
Hyndman, R. J., Wang, E., & Laptev, N. (2015). Large-scale unusual time series detection. In
Scrucca (br000150) 2012; 53
Goerg, G. M. (2014). ForeCA: An R package for Forecastable Component Analysis. R package version 0.1. URL
Fulcher, Jones (br000045) 2014; 26
Hyndman, Khandakar (br000085) 2008; 26
URL
Bandt, Pompe (br000010) 2002; 88
Clements, Hendry (br000020) 2001; 17
Goerg, G. M. (2013). Forecastable component analysis. In
Wang, Smith, Hyndman (br000165) 2006; 13
Petropoulos, Makridakis, Assimakopoulos, Nikolopoulos (br000145) 2014; 237
Zaccarelli, Li, Petrosillo, Zurlini (br000180) 2013; 28
Kang, Belušić, Smith-Miles (br000105) 2014; 71
Box, Cox (br000015) 1964; 26
Deng, Runger, Tuv, Vladimir (br000030) 2013; 239
Smith-Miles, Baatar, Wreford, Lewis (br000155) 2014; 45
Wolpert, Macready (br000175) 1997; 1
Fiorucci, J. A., Louzada, F., & Yiqi, B. (2016). forecTheta: Forecasting Time Series by Theta Models. R package version 2.2. URL
Kang, Belušić, Smith-Miles (br000110) 2015; 141
Makridakis, Hibon (br000125) 2000; 16
Lawrence (br000115) 2001; 17
Ord (br000140) 2001; 17
Mörchen (br000130) 2003
Wolpert (br000170) 1996; 8
.
Hyndman, Koehler (br000090) 2006; 22
Fulcher, Little, Jones (br000050) 2013; 10
Garland, James, Bradley (br000055) 2014; 90
Hyndman, R. J. (2013). Mcomp: Data from the M-competitions. R package version 2.05. URL
Fadlallah, Chen, Keil, Príncipe (br000035) 2013; 87
Smith-Miles, Bowly (br000160) 2015; 63
Maasoumi, Racine (br000120) 2002; 107
Hyndman, R. J. (2016). forecast: Forecasting functions for time series and linear models. R package version 7.1. URL
Nanopoulos, Alcock, Manolopoulos (br000135) 2001; 10
(pp. 64–72).
Cleveland, Cleveland, McRae, Terpenning (br000025) 1990; 6
Deng (10.1016/j.ijforecast.2016.09.004_br000030) 2013; 239
Garland (10.1016/j.ijforecast.2016.09.004_br000055) 2014; 90
Fulcher (10.1016/j.ijforecast.2016.09.004_br000045) 2014; 26
Box (10.1016/j.ijforecast.2016.09.004_br000015) 1964; 26
Wang (10.1016/j.ijforecast.2016.09.004_br000165) 2006; 13
Zaccarelli (10.1016/j.ijforecast.2016.09.004_br000180) 2013; 28
Cleveland (10.1016/j.ijforecast.2016.09.004_br000025) 1990; 6
Maasoumi (10.1016/j.ijforecast.2016.09.004_br000120) 2002; 107
Wolpert (10.1016/j.ijforecast.2016.09.004_br000170) 1996; 8
Bandt (10.1016/j.ijforecast.2016.09.004_br000010) 2002; 88
10.1016/j.ijforecast.2016.09.004_br000100
Assimakopoulos (10.1016/j.ijforecast.2016.09.004_br000005) 2000; 16
10.1016/j.ijforecast.2016.09.004_br000065
10.1016/j.ijforecast.2016.09.004_br000060
Hyndman (10.1016/j.ijforecast.2016.09.004_br000095) 2002; 18
Clements (10.1016/j.ijforecast.2016.09.004_br000020) 2001; 17
10.1016/j.ijforecast.2016.09.004_br000040
Nanopoulos (10.1016/j.ijforecast.2016.09.004_br000135) 2001; 10
Smith-Miles (10.1016/j.ijforecast.2016.09.004_br000155) 2014; 45
Petropoulos (10.1016/j.ijforecast.2016.09.004_br000145) 2014; 237
Scrucca (10.1016/j.ijforecast.2016.09.004_br000150) 2012; 53
Kang (10.1016/j.ijforecast.2016.09.004_br000110) 2015; 141
Lawrence (10.1016/j.ijforecast.2016.09.004_br000115) 2001; 17
Hyndman (10.1016/j.ijforecast.2016.09.004_br000080) 2014
Fadlallah (10.1016/j.ijforecast.2016.09.004_br000035) 2013; 87
Mörchen (10.1016/j.ijforecast.2016.09.004_br000130) 2003
Ord (10.1016/j.ijforecast.2016.09.004_br000140) 2001; 17
Wolpert (10.1016/j.ijforecast.2016.09.004_br000175) 1997; 1
Hyndman (10.1016/j.ijforecast.2016.09.004_br000090) 2006; 22
Hyndman (10.1016/j.ijforecast.2016.09.004_br000085) 2008; 26
10.1016/j.ijforecast.2016.09.004_br000075
Makridakis (10.1016/j.ijforecast.2016.09.004_br000125) 2000; 16
Fulcher (10.1016/j.ijforecast.2016.09.004_br000050) 2013; 10
Smith-Miles (10.1016/j.ijforecast.2016.09.004_br000160) 2015; 63
10.1016/j.ijforecast.2016.09.004_br000070
Kang (10.1016/j.ijforecast.2016.09.004_br000105) 2014; 71
References_xml – reference: Goerg, G. M. (2014). ForeCA: An R package for Forecastable Component Analysis. R package version 0.1. URL:
– volume: 71
  start-page: 1090
  year: 2014
  end-page: 1104
  ident: br000105
  article-title: Detecting and classifying events in noisy time series
  publication-title: Journal of the Atmospheric Sciences
– volume: 16
  start-page: 521
  year: 2000
  end-page: 530
  ident: br000005
  article-title: The theta model: a decomposition approach to forecasting
  publication-title: International Journal of Forecasting
– volume: 26
  start-page: 1
  year: 2008
  end-page: 22
  ident: br000085
  article-title: Automatic time series forecasting: the forecast package for R
  publication-title: Journal of Statistical Software
– volume: 17
  start-page: 574
  year: 2001
  end-page: 575
  ident: br000115
  article-title: Commentaries on the M3-Competition. Why another study?
  publication-title: International Journal of Forecasting
– volume: 88
  year: 2002
  ident: br000010
  article-title: Permutation entropy: A natural complexity measure for time series
  publication-title: Physical Review Letters
– year: 2003
  ident: br000130
  article-title: Time series feature extraction for data mining using DWT and DFT, Technical Report 33
– reference: Hyndman, R. J. (2013). Mcomp: Data from the M-competitions. R package version 2.05. URL:
– volume: 17
  start-page: 550
  year: 2001
  end-page: 554
  ident: br000020
  article-title: Explaining the results of the M3 forecasting competition
  publication-title: International Journal of Forecasting
– volume: 18
  start-page: 439
  year: 2002
  end-page: 454
  ident: br000095
  article-title: A state space framework for automatic forecasting using exponential smoothing methods
  publication-title: International Journal of Forecasting
– volume: 16
  start-page: 451
  year: 2000
  end-page: 476
  ident: br000125
  article-title: The M3-Competition: results, conclusions and implications
  publication-title: International Journal of Forecasting
– reference: URL:
– volume: 28
  start-page: 22
  year: 2013
  end-page: 30
  ident: br000180
  article-title: Order and disorder in ecological time-series: Introducing normalized spectral entropy
  publication-title: Ecological Indicators
– volume: 1
  start-page: 67
  year: 1997
  end-page: 82
  ident: br000175
  article-title: No free lunch theorems for optimization
  publication-title: IEEE Transactions on Evolutionary Computation
– volume: 239
  start-page: 142
  year: 2013
  end-page: 153
  ident: br000030
  article-title: A time series forest for classification and feature extraction
  publication-title: Information Sciences
– reference: Goerg, G. M. (2013). Forecastable component analysis. In
– volume: 6
  start-page: 3
  year: 1990
  end-page: 73
  ident: br000025
  article-title: STL: A seasonal-trend decomposition procedure based on loess
  publication-title: Journal of Official Statistics
– reference: Fiorucci, J. A., Louzada, F., & Yiqi, B. (2016). forecTheta: Forecasting Time Series by Theta Models. R package version 2.2. URL:
– volume: 53
  year: 2012
  ident: br000150
  article-title: GA: a package for genetic algorithms in R
  publication-title: Journal of Statistical Software
– volume: 13
  start-page: 335
  year: 2006
  end-page: 364
  ident: br000165
  article-title: Characteristic-based clustering for time series data
  publication-title: Data Mining and Knowledge Discovery
– volume: 90
  year: 2014
  ident: br000055
  article-title: Model-free quantification of time-series predictability
  publication-title: Physical Review E
– volume: 26
  start-page: 3026
  year: 2014
  end-page: 3037
  ident: br000045
  article-title: Highly comparative feature-based time-series classification
  publication-title: IEEE Transactions on Knowledge and Data Engineering
– volume: 26
  start-page: 211
  year: 1964
  end-page: 252
  ident: br000015
  article-title: An analysis of transformations
  publication-title: Journal of the Royal Statistical Society: Series B
– volume: 8
  start-page: 1341
  year: 1996
  end-page: 1390
  ident: br000170
  article-title: The lack of a priori distinctions between learning algorithms
  publication-title: Neural Computing
– reference: (pp. 64–72).
– reference: Hyndman, R. J. (2016). forecast: Forecasting functions for time series and linear models. R package version 7.1. URL:
– reference: .
– volume: 10
  start-page: 20130048
  year: 2013
  ident: br000050
  article-title: Highly comparative time-series analysis: the empirical structure of time series and their methods
  publication-title: Journal of the Royal Society Interface
– year: 2014
  ident: br000080
  article-title: Forecasting: principles and practice
– volume: 45
  start-page: 12
  year: 2014
  end-page: 24
  ident: br000155
  article-title: Towards objective measures of algorithm performance across instance space
  publication-title: Computers & Operations Research
– volume: 17
  start-page: 537
  year: 2001
  end-page: 541
  ident: br000140
  article-title: Commentaries on the M3-Competition. An introduction, some comments and a scorecard
  publication-title: International Journal of Forecasting
– volume: 141
  start-page: 2057
  year: 2015
  end-page: 2069
  ident: br000110
  article-title: Classes of structures in the stable atmospheric boundary layer
  publication-title: Quarterly Journal of the Royal Meteorological Society
– volume: 107
  start-page: 291
  year: 2002
  end-page: 312
  ident: br000120
  article-title: Entropy and predictability of stock market returns
  publication-title: Journal of Econometrics
– volume: 63
  start-page: 102
  year: 2015
  end-page: 113
  ident: br000160
  article-title: Generating new test instances by evolving in instance space
  publication-title: Computers & Operations Research
– volume: 87
  year: 2013
  ident: br000035
  article-title: Weighted-permutation entropy: A complexity measure for time series incorporating amplitude information
  publication-title: Physical Review E
– reference: Hyndman, R. J., Wang, E., & Laptev, N. (2015). Large-scale unusual time series detection. In
– volume: 237
  start-page: 152
  year: 2014
  end-page: 163
  ident: br000145
  article-title: ‘Horses for courses’ in demand forecasting
  publication-title: European Journal of Operational Research
– volume: 22
  start-page: 679
  year: 2006
  end-page: 688
  ident: br000090
  article-title: Another look at measures of forecast accuracy
  publication-title: International Journal of Forecasting
– volume: 10
  year: 2001
  ident: br000135
  article-title: Feature-based classification of time-series data
  publication-title: International Journal of Computer Research
– volume: 107
  start-page: 291
  issue: 1–2
  year: 2002
  ident: 10.1016/j.ijforecast.2016.09.004_br000120
  article-title: Entropy and predictability of stock market returns
  publication-title: Journal of Econometrics
  doi: 10.1016/S0304-4076(01)00125-7
– volume: 239
  start-page: 142
  year: 2013
  ident: 10.1016/j.ijforecast.2016.09.004_br000030
  article-title: A time series forest for classification and feature extraction
  publication-title: Information Sciences
  doi: 10.1016/j.ins.2013.02.030
– year: 2014
  ident: 10.1016/j.ijforecast.2016.09.004_br000080
– ident: 10.1016/j.ijforecast.2016.09.004_br000060
– volume: 16
  start-page: 451
  issue: 4
  year: 2000
  ident: 10.1016/j.ijforecast.2016.09.004_br000125
  article-title: The M3-Competition: results, conclusions and implications
  publication-title: International Journal of Forecasting
  doi: 10.1016/S0169-2070(00)00057-1
– ident: 10.1016/j.ijforecast.2016.09.004_br000100
– volume: 1
  start-page: 67
  issue: 1
  year: 1997
  ident: 10.1016/j.ijforecast.2016.09.004_br000175
  article-title: No free lunch theorems for optimization
  publication-title: IEEE Transactions on Evolutionary Computation
  doi: 10.1109/4235.585893
– volume: 26
  start-page: 211
  issue: 2
  year: 1964
  ident: 10.1016/j.ijforecast.2016.09.004_br000015
  article-title: An analysis of transformations
  publication-title: Journal of the Royal Statistical Society: Series B
  doi: 10.1111/j.2517-6161.1964.tb00553.x
– ident: 10.1016/j.ijforecast.2016.09.004_br000040
– volume: 90
  issue: 5
  year: 2014
  ident: 10.1016/j.ijforecast.2016.09.004_br000055
  article-title: Model-free quantification of time-series predictability
  publication-title: Physical Review E
  doi: 10.1103/PhysRevE.90.052910
– volume: 45
  start-page: 12
  issue: 0
  year: 2014
  ident: 10.1016/j.ijforecast.2016.09.004_br000155
  article-title: Towards objective measures of algorithm performance across instance space
  publication-title: Computers & Operations Research
  doi: 10.1016/j.cor.2013.11.015
– volume: 17
  start-page: 537
  year: 2001
  ident: 10.1016/j.ijforecast.2016.09.004_br000140
  article-title: Commentaries on the M3-Competition. An introduction, some comments and a scorecard
  publication-title: International Journal of Forecasting
– ident: 10.1016/j.ijforecast.2016.09.004_br000075
– volume: 71
  start-page: 1090
  issue: 3
  year: 2014
  ident: 10.1016/j.ijforecast.2016.09.004_br000105
  article-title: Detecting and classifying events in noisy time series
  publication-title: Journal of the Atmospheric Sciences
  doi: 10.1175/JAS-D-13-0182.1
– volume: 88
  year: 2002
  ident: 10.1016/j.ijforecast.2016.09.004_br000010
  article-title: Permutation entropy: A natural complexity measure for time series
  publication-title: Physical Review Letters
  doi: 10.1103/PhysRevLett.88.174102
– volume: 17
  start-page: 550
  year: 2001
  ident: 10.1016/j.ijforecast.2016.09.004_br000020
  article-title: Explaining the results of the M3 forecasting competition
  publication-title: International Journal of Forecasting
– volume: 18
  start-page: 439
  issue: 3
  year: 2002
  ident: 10.1016/j.ijforecast.2016.09.004_br000095
  article-title: A state space framework for automatic forecasting using exponential smoothing methods
  publication-title: International Journal of Forecasting
  doi: 10.1016/S0169-2070(01)00110-8
– volume: 28
  start-page: 22
  year: 2013
  ident: 10.1016/j.ijforecast.2016.09.004_br000180
  article-title: Order and disorder in ecological time-series: Introducing normalized spectral entropy
  publication-title: Ecological Indicators
  doi: 10.1016/j.ecolind.2011.07.008
– volume: 53
  issue: 4
  year: 2012
  ident: 10.1016/j.ijforecast.2016.09.004_br000150
  article-title: GA: a package for genetic algorithms in R
  publication-title: Journal of Statistical Software
  doi: 10.18637/jss.v053.i04
– volume: 87
  year: 2013
  ident: 10.1016/j.ijforecast.2016.09.004_br000035
  article-title: Weighted-permutation entropy: A complexity measure for time series incorporating amplitude information
  publication-title: Physical Review E
  doi: 10.1103/PhysRevE.87.022911
– volume: 16
  start-page: 521
  issue: 4
  year: 2000
  ident: 10.1016/j.ijforecast.2016.09.004_br000005
  article-title: The theta model: a decomposition approach to forecasting
  publication-title: International Journal of Forecasting
  doi: 10.1016/S0169-2070(00)00066-2
– volume: 6
  start-page: 3
  issue: 1
  year: 1990
  ident: 10.1016/j.ijforecast.2016.09.004_br000025
  article-title: STL: A seasonal-trend decomposition procedure based on loess
  publication-title: Journal of Official Statistics
– volume: 22
  start-page: 679
  issue: 4
  year: 2006
  ident: 10.1016/j.ijforecast.2016.09.004_br000090
  article-title: Another look at measures of forecast accuracy
  publication-title: International Journal of Forecasting
  doi: 10.1016/j.ijforecast.2006.03.001
– volume: 237
  start-page: 152
  issue: 1
  year: 2014
  ident: 10.1016/j.ijforecast.2016.09.004_br000145
  article-title: ‘Horses for courses’ in demand forecasting
  publication-title: European Journal of Operational Research
  doi: 10.1016/j.ejor.2014.02.036
– ident: 10.1016/j.ijforecast.2016.09.004_br000065
– volume: 17
  start-page: 574
  year: 2001
  ident: 10.1016/j.ijforecast.2016.09.004_br000115
  article-title: Commentaries on the M3-Competition. Why another study?
  publication-title: International Journal of Forecasting
– volume: 10
  start-page: 20130048
  issue: 83
  year: 2013
  ident: 10.1016/j.ijforecast.2016.09.004_br000050
  article-title: Highly comparative time-series analysis: the empirical structure of time series and their methods
  publication-title: Journal of the Royal Society Interface
  doi: 10.1098/rsif.2013.0048
– year: 2003
  ident: 10.1016/j.ijforecast.2016.09.004_br000130
– volume: 13
  start-page: 335
  issue: 3
  year: 2006
  ident: 10.1016/j.ijforecast.2016.09.004_br000165
  article-title: Characteristic-based clustering for time series data
  publication-title: Data Mining and Knowledge Discovery
  doi: 10.1007/s10618-005-0039-x
– volume: 26
  start-page: 3026
  issue: 12
  year: 2014
  ident: 10.1016/j.ijforecast.2016.09.004_br000045
  article-title: Highly comparative feature-based time-series classification
  publication-title: IEEE Transactions on Knowledge and Data Engineering
  doi: 10.1109/TKDE.2014.2316504
– volume: 26
  start-page: 1
  issue: 3
  year: 2008
  ident: 10.1016/j.ijforecast.2016.09.004_br000085
  article-title: Automatic time series forecasting: the forecast package for R
  publication-title: Journal of Statistical Software
– volume: 141
  start-page: 2057
  issue: 691
  year: 2015
  ident: 10.1016/j.ijforecast.2016.09.004_br000110
  article-title: Classes of structures in the stable atmospheric boundary layer
  publication-title: Quarterly Journal of the Royal Meteorological Society
  doi: 10.1002/qj.2501
– volume: 63
  start-page: 102
  year: 2015
  ident: 10.1016/j.ijforecast.2016.09.004_br000160
  article-title: Generating new test instances by evolving in instance space
  publication-title: Computers & Operations Research
  doi: 10.1016/j.cor.2015.04.022
– volume: 10
  issue: 3
  year: 2001
  ident: 10.1016/j.ijforecast.2016.09.004_br000135
  article-title: Feature-based classification of time-series data
  publication-title: International Journal of Computer Research
– volume: 8
  start-page: 1341
  issue: 7
  year: 1996
  ident: 10.1016/j.ijforecast.2016.09.004_br000170
  article-title: The lack of a priori distinctions between learning algorithms
  publication-title: Neural Computing
  doi: 10.1162/neco.1996.8.7.1341
– ident: 10.1016/j.ijforecast.2016.09.004_br000070
SSID ssj0005711
Score 2.544532
Snippet It is common practice to evaluate the strength of forecasting methods using collections of well-studied time series datasets, such as the M3 data. The question...
SourceID crossref
elsevier
SourceType Enrichment Source
Index Database
Publisher
StartPage 345
SubjectTerms Forecasting algorithm comparison
M3-Competition
Time series generation
Time series visualisation
Title Visualising forecasting algorithm performance using time series instance spaces
URI https://dx.doi.org/10.1016/j.ijforecast.2016.09.004
Volume 33
WOSCitedRecordID wos000399512200002&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  customDbUrl:
  eissn: 1872-8200
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0005711
  issn: 0169-2070
  databaseCode: AIEXJ
  dateStart: 19950301
  isFulltext: true
  titleUrlDefault: https://www.sciencedirect.com
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1db9MwFLXKxgMvE-NDdAPkBx5AUyrXzoctniY0NIYYCAYqT5Ht2JCqS6umm8Zf4FdzHTtpxiYxhHipqqhuHN-T6-urc89F6BkV1soMTqpK2TiKTUIiCXFDZFLNScEzJahsmk1kx8d8MhEfBoOfbS3M-SyrKn5xIRb_1dRwDYztSmf_wtzdn8IF-A5Gh08wO3zeyPBfytoVStaBImm0rBtqs5x9my_L1fdTJ1XcFQucNb9zHeb33OwaepYLGJ3y7MLRtfrR6-X0YU90onefzoOHRPRXWVlTduj5URUh5_pxrvaORpcyPNG7MrTgfisDASkkJGCTW_NYmizZlUoZn7hMBZjJNwkZGe9seQbe2CuVdt7Yy2IE1NGea2VedjLs0swLvl_ZAHwuYjoqp-2TO_pe2mjZ-j7Hv8lrf3IzcxODwNT1vCC30CbNEgEecnP_zcHkaE0Yysa-u2V4ksAL82zB6-93fbDTC2BO7qKtcPLA-x4x22hgqnto6MuzcXDxNX4edMhf3Efve0jCPQvjDkm4hyTcIAk7JGGPJNwiCXskPUCfXx-cvDqMQv-NSDPOVpHirDDEWgMvLeWEmrGNSREzrpJYyEJpoRJrTApnUpakTMH62SKJM01ilcKxgz1EG9W8Mo8QVk4DnlOtqUpjC_OVklLBU0WUSGC_HaKsXahcB3F61yNllrcsxGm-XuLcLXFORA5LPETjbuTCC7TcYMzL1hZ5CDR9AJkDjP44euefRu-iO-uX5jHaWC3PzBN0W5-vynr5NGDuFx95rMs
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Visualising+forecasting+algorithm+performance+using+time+series+instance+spaces&rft.jtitle=International+journal+of+forecasting&rft.au=Kang%2C+Yanfei&rft.au=Hyndman%2C+Rob+J.&rft.au=Smith-Miles%2C+Kate&rft.date=2017-04-01&rft.pub=Elsevier+B.V&rft.issn=0169-2070&rft.eissn=1872-8200&rft.volume=33&rft.issue=2&rft.spage=345&rft.epage=358&rft_id=info:doi/10.1016%2Fj.ijforecast.2016.09.004&rft.externalDocID=S0169207016301030
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0169-2070&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0169-2070&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0169-2070&client=summon