옮긴이머리말xiii
베타리더후기xiv
머리말xvi
감사의말xviii
이책에대하여xix
표지에대하여xxiii
PARTI시간은그누구도기다려주지않는다
CHAPTER1시계열예측의이해3
1.1시계열소개4
__1.1.1시계열의구성요소5
1.2시계열예측에대한조감도8
__1.2.1목적설정하기10/1.2.2목적을달성하기위해무엇을예측해야하는지결정하기10/1.2.3예측할기간설정하기10/1.2.4데이터수집하기10/1.2.5예측모델개발하기11/1.2.6상용환경에배포하기12/1.2.7모니터링하기12/1.2.8새로운데이터수집하기12
1.3시계열예측이다른회귀작업들과다른점13
__1.3.1시계열에는순서가있다13/1.3.2시계열에특징이없는경우가있다14
1.4다음단계14
요약15
CHAPTER2단순하게미래예측하기16
2.1베이스라인모델정의하기18
2.2과거평균으로예측하기19
__2.2.1베이스라인구현을위한설정20/2.2.2과거평균기반베이스라인모델구현하기22
2.3작년의평균으로예측하기27
2.4마지막으로측정된값으로예측하기29
2.5단순한계절적예측구현하기31
2.6다음단계32
요약33
CHAPTER3확률보행따라가보기35
3.1확률보행프로세스37
__3.1.1확률보행프로세스시뮬레이션하기37
3.2확률보행식별하기40
__3.2.1정상성42/3.2.2정상성테스트하기44/3.2.3자기상관함수48/3.2.4모든것을종합하기48/3.2.5GOOGL은확률보행인가?52
3.3확률보행예측하기55
__3.3.1긴기간예측하기55/3.3.2다음시간단계예측하기61
3.4다음단계64
3.5연습65
__3.5.1확률보행시뮬레이션하기와예측하기65/3.5.2GOOGL의일일종가예측하기66/3.5.3직접선택한종목의일일종가예측하기66
요약67
PART2통계적모델을사용하여예측하기
CHAPTER4이동평균과정모델링하기71
4.1이동평균과정정의하기73
__4.1.1이동평균과정의차수식별하기75
4.2이동평균과정예측하기80
4.3다음단계90
4.4연습91
__4.4.1MA(2)프로세스시뮬레이션하고예측수행하기92/4.4.2MA(q)프로세스시뮬레이션하고예측수행하기92
요약93
CHAPTER5자기회귀과정모델링하기94
5.1소매점의주간평균유동인구예측하기95
5.2자기회귀과정정의하기97
5.3정상적자기회귀과정의차수찾기98
__5.3.1편자기상관함수104
5.4자기회귀과정예측하기107
5.5다음단계114
5.6연습114
__5.6.1AR(2)프로세스시뮬레이션하고예측수행하기114/5.6.2AR(p)프로세스시뮬레이션하고예측수행하기115
요약115
CHAPTER6복잡한시계열모델링하기116
6.1데이터센터의대역폭사용량예측하기117
6.2자기회귀이동평균과정살펴보기120
6.3정상적ARMA프로세스식별하기122
6.4일반적모델링절차고안하기128
__6.4.1아카이케정보기준이해하기130/6.4.2AIC를사용하여모델선택하기132/6.4.3잔차분석이해하기134/6.4.4잔차분석수행하기139
6.5일반적모델링절차적용하기143
6.6대역폭사용량예측하기152
6.7다음단계157
6.8연습157
__6.8.1시뮬레이션된ARMA(1,1)프로세스에대한예측수행하기158/6.8.2ARMA(2,2)프로세스시뮬레이션하고예측수행하기158
요약159
CHAPTER7비정상적시계열예측하기161
7.1자기회귀누적이동평균모델정의하기164
7.2비정상적시계열에적용하기위해일반적모델링절차수정하기165
7.3비정상적시계열예측하기167
7.4다음단계177
7.5연습177
__7.5.14장,5장,6장의데이터집합에ARIMA(p,d,q)모델적용하기177
요약178
CHAPTER8계절성고려하기179
8.1SARIMA(p,d,q)(P,D,Q)m모델살펴보기180
8.2시계열에서계절별패턴식별하기183
8.3월간항공승객수예측하기187
__8.3.1ARIMA(p,d,q)모델을사용하여예측하기190/8.3.2SARIMA(p,d,q)(P,D,Q)m모델을사용하여예측하기196/8.3.3각예측방법의성능비교하기200
8.4다음단계203
8.5연습203
__8.5.1존슨앤드존슨데이터집합에SARIMA(p,d,q)(P,D,Q)m모델적용하기203
요약204
CHAPTER9모델에외생변수추가하기205
9.1SARIMAX모델살펴보기207
__9.1.1미국거시경제데이터집합의외생변수탐색하기208/9.1.2SARIMAX사용시유의사항211
9.2SARIMAX모델을사용하여실질GDP예측하기212
9.3다음단계221
9.4연습222
__9.4.1SARIMAX모델에모든외생변수를사용하여실질GDP예측하기222
요약222
CHAPTER10다중시계열예측하기223
10.1VAR모델살펴보기225
10.2VAR(p)모델에대한모델링절차설계하기227
__10.2.1그레인저인과관계테스트살펴보기229
10.3실질가처분소득과실질소비예측하기230
10.4다음단계242
10.5연습243
__10.5.1VARMA모델을사용하여realdpi와realcons예측하기243/10.5.2VARMAX모델을사용하여realdpi와realcons예측하기244
요약244
CHAPTER11캡스톤프로젝트:호주의항당뇨제처방건수예측하기245
11.1필요한라이브러리임포트하고데이터로딩하기247
11.2수열과그구성요소시각화하기248
11.3데이터로모델링하기250
__11.3.1모델선택수행하기253/11.3.2잔차분석수행하기254
11.4예측을수행하고,모델성능평가하기256
11.5다음단계260
PART3딥러닝을활용하여대규모예측하기
CHAPTER12시계열예측을위한딥러닝소개하기263
12.1시계열예측에딥러닝을사용해야하는경우264
12.2다양한유형의딥러닝모델살펴보기265
12.3예측을위한딥러닝적용준비하기268
__12.3.1데이터탐색수행하기268/12.3.2특징엔지니어링과데이터분할272
12.4다음단계277
12.5연습277
요약278
CHAPTER13딥러닝을위해데이터윈도잉하고베이스라인모델만들기279
13.1데이터윈도우만들기280
__13.1.1시계열예측을위한딥러닝모델을훈련하는방법살펴보기280
__13.1.2DataWindow클래스구현하기284
13.2베이스라인모델적용하기292
__13.2.1단일단계베이스라인모델292
__13.2.2다중단계베이스라인모델295
__13.2.3다중출력베이스라인모델299
13.3다음단계303
13.4연습303
요약304
CHAPTER14딥러닝첫걸음305
14.1선형모델구현하기306
__14.1.1단일단계선형모델구현하기307/14.1.2다중단계선형모델구현하기309/14.1.3다중출력선형모델구현하기311
14.2심층신경망구현하기312
__14.2.1단일단계모델로심층신경망구현하기314/14.2.2다중단계모델로심층신경망구현하기317/14.2.3다중출력모델로서심층신경망구현하기319
14.3다음단계320
14.4연습321
요약322
CHAPTER15LSTM으로과거기억하기323
15.1순환신경망살펴보기324
15.2LSTM아키텍처살펴보기326
__15.2.1망각게이트327/15.2.2입력게이트329/15.2.3출력게이트330
15.3LSTM아키텍처구현하기332
__15.3.1단일단계모델로서LSTM구현하기332/15.3.2다중단계모델로서LSTM구현하기335/15.3.3다중출력모델로서LSTM구현하기338
15.4다음단계341
15.5연습342
요약343
CHAPTER16CNN으로시계열필터링하기344
16.1CNN살펴보기345
16.2CNN구현하기349
__16.2.1CNN을단일단계모델로서구현하기350/16.2.2CNN을다중단계모델로서구현하기354/16.2.3CNN을다중출력모델로서구현하기356
16.3다음단계359
16.4연습359
요약361
CHAPTER17예측으로더많은예측하기362
17.1ARLSTM아키텍처살펴보기363
17.2자기회귀LSTM모델구축하기364
17.3다음단계370
17.4연습371
요약371
CHAPTER18캡스톤프로젝트:가정의전력소비량예측하기372
18.1캡스톤프로젝트이해하기373
__18.1.1캡스톤프로젝트의목표375
18.2데이터랭글링및전처리하기376
__18.2.1누락된데이터처리하기377/18.2.2데이터변환379/18.2.3데이터리샘플링하기379
18.3특징엔지니어링382
__18.3.1불필요한열제거하기383/18.3.2계절적기간식별하기383/18.3.3데이터를분할하고규모조정하기386
18.4딥러닝으로모델링할준비하기387
__18.4.1초기설정387/18.4.2DataWindow클래스정의하기389/18.4.3모델훈련을위한유틸리티함수391
18.5딥러닝으로모델링하기392
__18.5.1베이스라인모델392/18.5.2선형모델396/18.5.3심층신경망397/18.5.4장단기메모리모델398/18.5.5합성곱신경망399/18.5.6CNN과LSTM결합하기401/18.5.7자기회귀LSTM모델402/18.5.8최적의모델선택하기404
18.6다음단계406
PART4대규모예측자동화하기
CHAPTER19Prophet으로시계열예측자동화하기409
19.1자동화된예측라이브러리들에대한개관410
19.2Prophet살펴보기412
19.3Prophet을사용하여기본적예측해보기414
19.4Prophet의고급기능살펴보기420
__19.4.1시각화기능421/1
