## Forecasting at scale

✅ #今日の積み上げ #デイトラ Python 中級編 1~3 Prophetを用いた株価予測プログラムの作成 Prophetが凄すぎる......ッ！ けど一体何が起こってるのか分からない

@therealfuzzy2 @MKreutzfeldt Rtlive verwendet Prophet. Implementierung gibt’s für R und Python. Könnte im Prinzip auch hier für Trendanalysen nützlich sein. https://t.co/1DLdFS4vWi https://t.co/NIltokeWDe https://t.co/lD4FmRufW0
@SimonWMnhs I worked with it a bit and it's nice. We went with Holt-winters type smoothers, after great sessions on forecasting from @Bahman_R_T. @seanjtaylor 's paper on Proohet was interesting https://t.co/IiNJ3qRtQd
RT @PyDataJeddah: Then we looked into the basic idea and mathematics behind the Prophet library to know how #fbprophet can fit both the tre…
Forecasting at scale https://t.co/qY4sDm6Zfe via @PeerJPreprints
Then we looked into the basic idea and mathematics behind the Prophet library to know how #fbprophet can fit both the trend and the seasonality components for a time series. Check out this paper. https://t.co/VJwCWSuwaW
@mkeulemans Dit soort voorspellingen zou je ook kunnen doen met een algoritme als Prophet (bv https://t.co/QzVEAuiKIn). Maar met de puntschattingen van twee eenvoudige modellen een ruisloze voorspelling geven -- statistisch zet ik hier mijn vraagtekens bij.
うーん，fbprophetを状態空間モデルの実装であるという人がいますけれど，このstanの実装とhttps://t.co/X5mJ9MrhW9 Prophetクラスの実装 https://t.co/NElwfzI35O および説明書 https://t.co/VoG2eCyyS0 のどこをみれば状態空間モデルだといえるのでしょうかね…
Forecasting at scale a modular regression model with interpretable parameters that can be intuitively adjusted by analysts with domain knowledge about the time series https://t.co/FMQpgMMomZ
RT @bearloga: @fulhack • The paper on Prophet from @seanjtaylor &amp; @_bletham_: https://t.co/JRJ5ruLObW • Real-world application of causal im…
@fulhack • The paper on Prophet from @seanjtaylor &amp; @_bletham_: https://t.co/JRJ5ruLObW • Real-world application of causal impact methodology https://t.co/W7KN1A3frZ by @chelsyxie et al. • Time series models chapter in Stan user manual: https://t.co/JTbVGZ5ybz
prophetの元論文見つけたので、読み切った。メモを残す: https://t.co/lONQ0uG9Zb
386 days ago
@danielhfrank @gkossakowski @snoble I think you want to use @seanjtaylor's trick from 3.1.1 of https://t.co/y4dx1VIegg to vectorize the changepoints.
RT @PTKDaily:
RT @PTKDaily:
400 days ago
RT @PTKDaily:
414 days ago
RT @PTKDaily:
RT @PTKDaily:
422 days ago
Facebook'un çekirdek veri bilimi ekibi tarafından hazırlanan; açık kaynak, zaman serisi veri tahminleme (time series data forecasting) kütüphanesi Prophet'in yeni versiyonu (0.6) yayınlandı. Kod: https://t.co/rz6o4GOaud Makale: https://t.co/gEdYdFuMZw
@rob_rix Hierarchical modeling works for this. See for example @seanjtaylor 's Prophet writeup https://t.co/m1iHIKwZpq and @mcmc_stan implementation https://t.co/2vaS5Sx1gV
Prophetの論文読んでる 知識が全然無いから時間が読むのにかかるなぁ https://t.co/pYlQDLAJnl

NOT PEER-REVIEWED
"PeerJ Preprints" is a venue for early communication or feedback before peer review. Data may be preliminary.

### Competing Interests

Sean J Taylor and Benjamin Letham are both employees at Facebook, Inc.

### Author Contributions

Sean J Taylor analyzed the data, wrote the paper, prepared figures and/or tables, performed the computation work, reviewed drafts of the paper.

Benjamin Letham analyzed the data, wrote the paper, prepared figures and/or tables, performed the computation work, reviewed drafts of the paper.

### Data Deposition

The following information was supplied regarding data availability:

All code is available in our open source repository: https://github.com/facebookincubator/prophet

### Funding

The authors received no funding for this work.

Some Markdown syntax is allowed: _italic_ **bold** ^superscript^ ~subscript~ %%blockquote%% [link text](link URL)