TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Time Series Prediction – A short introduction for pragmatists

393 pointsby makaimcover 5 years ago

12 comments

wjncover 5 years ago
And now for the great thing ... Prophet uses Stan underneath [1] and thus is built on foundations of &#x27;regular&#x27; Bayesian statistics. Andrew Gelman has written about Prophet as well [2].<p>After reading this blog I am tempted to get the ML for time series book though. I&#x27;d love to try and compare some less than trivial examples with covariates involved.<p>[1] <a href="https:&#x2F;&#x2F;peerj.com&#x2F;preprints&#x2F;3190&#x2F;" rel="nofollow">https:&#x2F;&#x2F;peerj.com&#x2F;preprints&#x2F;3190&#x2F;</a><p>[2] <a href="https:&#x2F;&#x2F;statmodeling.stat.columbia.edu&#x2F;2017&#x2F;03&#x2F;01&#x2F;facebooks-prophet-uses-stan&#x2F;" rel="nofollow">https:&#x2F;&#x2F;statmodeling.stat.columbia.edu&#x2F;2017&#x2F;03&#x2F;01&#x2F;facebooks-...</a>
评论 #21494034 未加载
评论 #21494650 未加载
评论 #21494339 未加载
chroem-over 5 years ago
Basis function regression is a very under-appreciated method for producing time series forecasts. I&#x27;ve found that it beats most of the methods described in this article. Maybe I should make a blog post...
评论 #21496033 未加载
评论 #21492927 未加载
platzover 5 years ago
Normally the method for dealing with &#x27;time series&#x27; is really just finding ways to turn a non-stationary distribution into a stationary distribution, where you can then apply classic statistical methods on them. So you&#x27;re just finding ways to factor out the time component in the data so you can use the standard non-time sensitive regression models on the transformed data.<p>I don&#x27;t think it&#x27;s untill you get to the NN based models that they start treating time as a first-class component in the model.<p>* If I&#x27;m wrong please explain why instead of downvoting
评论 #21497429 未加载
评论 #21502020 未加载
评论 #21499336 未加载
graycatover 5 years ago
Here are some old references for the problem of the OP:<p>David R. Brillinger, <i>Time Series Analysis: Data Analysis and Theory, Expanded Edition</i>, ISBN 0-8162-1150-7, Holden-Day, San Francisco, 1981.<p>George E. P. Box and Gwilym M. Jenkins, <i>Time Series Analysis --- Forecasting and Control: Revised Edition</i>, ISBN 0-8162-1104-3, Holden-Day, San Francisco, 1976.<p>Brillinger was a John Tukey student at Princeton and long at Berkeley.
hprotagonistover 5 years ago
My very first move with timeseries data is to get to frequency space as fast as I can.
评论 #21495101 未加载
评论 #21494842 未加载
评论 #21492968 未加载
评论 #21493105 未加载
_v7guover 5 years ago
I mostly feel these methods are quite overkill for most applications. As a purist, I&#x27;d recommend starting out with a simple linear regression and then moving on to adding methods to cover the letters of SARIMA by showing the need for each. It may not be as flashy, but linear regression is a stupidly powerful and very cheap tool for all kinds of situations.
评论 #21501962 未加载
cube2222over 5 years ago
I really recommend Prophet as an easy to use option like the article says.<p>I needed anomaly detection for prometheus metrics integrated with grafana for marking &quot;anomalous&quot; regions so the model doesn&#x27;t learn them.<p>Took me a week to set it all up including packaging up as a Microservice and deploying.
ayayecocojamboover 5 years ago
In the multivariate time series forecasting problem i found out that fine-tuned xgboost (and its variants) performs much better than fbprophet, sarimax, RNN variations. Predicting time series with RNN is like killing a bird for bazooka.
oli5679over 5 years ago
I really appreciate the philosophy of defining metric, and measuring performance going from simple to complex methods.<p>I&#x27;ve also used Prophet library, and find it works well out of the box.
TrackerFFover 5 years ago
Gaussian Processes seems to be pretty popular, might want to include that to the comparisons.
评论 #21493693 未加载
jlduggerover 5 years ago
Anyone patched prophet into graphite? Curious if the two are easily combined.
person_of_colorover 5 years ago
This is one type of time series, but how about audio time series?