ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

(1/2) Microsoft Forecasting Framework for Finance Time Series 🚀

The finnts is an R package from Microsoft for automated forecasting for financial forecast. This includes the following features:
✅ Automated feature engineering and feature selection
✅ Training models with backtesting, scoring, and ranking
✅ Supports more than 25 models for univariate and multivariate time series
✅ Supports daily, weekly, monthly, quarterly, and yearly frequency

ramikrispin, (edited )
@ramikrispin@mstdn.social avatar

Installation: install.packages("finnts")

(2/2) License 🪪: MIT 🦄

Code 🔗: https://github.com/microsoft/finnts/
Documentation 📖: https://microsoft.github.io/finnts/

Thanks to the authors - Mike Tokic and Aadharsh Kannan! 🙏🏼

nrennie, to random
@nrennie@fosstodon.org avatar

Thanks to Forecasting for Social Good for inviting me to deliver a workshop on "Forecasting with Generalised Additive Models (GAMs) in R" today! We talked about:

❓ What are GAMs?
❓ How do we fit GAMs in R?
❓ Model evaluation and forecasting
❓ When GAMs go wrong!

Slides: https://nrennie.github.io/f4sg-gams

GitHub: https://github.com/nrennie/f4sg-gams

Posit Cloud: https://posit.cloud/content/7637971

nrennie,
@nrennie@fosstodon.org avatar

And a blog post answering some of the questions that we ran out of time for!

https://nrennie.rbind.io/blog/forecasting-gams-r-questions/

ramikrispin, to python
@ramikrispin@mstdn.social avatar

(1/3)Modeling Short Time Series with Prior Knowledge in PyMC 🚀

Yesterday, I shared an article by Tim Radtke about forecasting insufficient time series data with a Bayesian approach using R. Here is the Python version 🧵👇🏼

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/3) The TLDR is when you need to model a short time series (less than one seasonal cycle) and have some knowledge or assumption about the expected behavior of the series - either from a similar series (i.e., similar products or geos) you can translate those assumptions to the model's prior distributions and use it to build a forecasting model.

ramikrispin,
@ramikrispin@mstdn.social avatar

(3/3) Here is the Python version of this approach by Juan Camilo Orduz using Python 🐍 with PyMC:
https://juanitorduz.github.io/short_time_series_pymc/

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

(1/2) Modeling Short Time Series with Prior Knowledge

When modeling time series data, you may find yourself with insufficient data. Insufficient data in time series would typically be defined as less than one seasonal cycle. This would challenge us to understand whether some events are driven by seasonality or other reasons, such as one-time events, outliers, etc.

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/2) If you have some prior knowledge about the series (e.g., learning from similar products or goes, etc.), you should consider using the Bayesian approach.

The article below by Tim Radtke provides an example of how to incorporate prior assumptions into a time series forecasting model when having insufficient data.

https://minimizeregret.com/short-time-series-prior-knowledge

JoshuaHolland, to Weather
@JoshuaHolland@mastodon.social avatar

Weird day. We were expected to get 6-10 inches of snow starting at 2 am and lasting through this afternoon. Schools closed. City declared a snow emergency. But it just started snowing at 8:30 and it looks like it'll only be a dusting.

Hudson Valley : “This is one of the more severe changes [to a forecast] we’ve seen in recent memory. For the data to be unanimous in the result… and then to all begin shifting dramatically SE in about 12 to 18 hours, is a rare event.”

nrennie, to random
@nrennie@fosstodon.org avatar

Join me on February 21st for a Forecasting for Social Good (F4SG) workshop about forecasting using GAMs in R! 📈

Register here: https://cardiff.zoom.us/meeting/register/tZEqduGsqjwqGNWoG7FSA6U51ohnswp74Ww-#/registration

ramikrispin, to machinelearning
@ramikrispin@mstdn.social avatar

(1/3) My talk from the Øredev 2023 conference about forecasting at scale is now available 👇🏼

📽️: https://www.youtube.com/watch?v=pUvVfO0_Kmc

#timeseries #forecasting #rstats #machinelearning #DataScience #oredev

ramikrispin,
@ramikrispin@mstdn.social avatar

(2/3) The talk focuses on methods to identify patterns in time series using cluster analysis inspired by Prof Rob Hyndman's paper about Feature-based time series analysis 👇🏼
Seminar 🔗: https://robjhyndman.com/seminars/fbtsa-ssc/

ramikrispin,
@ramikrispin@mstdn.social avatar

(3/3) Resources 📚
Slides: https://github.com/RamiKrispin/talks/tree/main/oredev%202023-%20Forecasting%20at%20Scale
Code (⚠️not documented yet⚠️): https://github.com/RamiKrispin/forecasting-at-scale

Thank you to the conference organizers for the invite! 🙏🏼

ai6yr, (edited ) to LosAngeles
@ai6yr@m.ai6yr.org avatar

Radar at 12/21 1804 currently shows the storm is parked offshore from Los Angeles (dumping quite a bit there). Still raining in Oxnard.

ai6yr,
@ai6yr@m.ai6yr.org avatar

So, the ghost in the HRRR machine puts totals near Santa Barbara (on 12/22/23 at 9am) at less than an Inch. But at 12/22/23 at 10am then says it thinks total accumulation will be 5 inches. It's likely a glitch (the HRRR accumulation algorithm), but I do wonder what's going to be correct.

image/png

jeffowski, to ai
@jeffowski@mastodon.world avatar
_9CL7T9k8cjnD_,
Daojoan, to random
@Daojoan@mastodon.social avatar

Tech workers who make 6-7 figures a year destroyed the income of freelancers earning pennies and we called it progress.

_9CL7T9k8cjnD_,

@Daojoan Just wait. It's going to get MUCH WORSE. Anything that is rule driven is in the : are coming to the end of their

image/gif
image/gif

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

It was a pleasure to present about forecasting at scale at the South Dakota State University Data Science Club meetup last month. The event recording is now available 👇🏼

Video 📽️: https://www.youtube.com/watch?v=_T1TVME9vNc
Slides 📖: https://github.com/RamiKrispin/talks/tree/main/202310%20SDSU%20Data%20Science%20Club%20-%20Forecasting%20at%20Scale

Thanks to Prof Steven Ge for the invite! 🙏🏼

appassionato, to bookstodon
@appassionato@mastodon.social avatar

The Collapse of Western Civilization: A View From the Future

This work is an important title that will change how readers look at the world. Dramatizing climate change in ways traditional nonfiction cannot, this inventive, at times humorous work reasserts the importance of scientists and the work they do and reveals the self-serving interests of the so called carbon industrial complex that have turned the practice of sound science into political fodder.

@bookstodon

Nonog, to random

Disaster early-warning systems are ‘doomed to fail’ — only collective action can plug the gaps
From floods to wildfires, and tsunamis to volcanic eruptions, early-warning systems can stop natural hazards becoming human disasters. But more joined-up thinking is urgently needed.
https://www.nature.com/articles/d41586-023-03510-8 #disaster #EarlyWarning #systems #NaturalHazards #forecasting

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

Working on a talk focusing on methods for analyzing and forecasting time series at scale. Having fun with Prof. Rob Hyndman's approach to analyzing multiple time series with dimension reduction for time series features with principle components 😍

Code (WIP) 🔗: https://github.com/RamiKrispin/forecasting-at-scale

video/mp4

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

I am excited to present today at the SDSU Data Science Club meetup on approaches for forecasting at scale. The event at 4 PM PST / 6 PM CDT is open to the public via Zoom, and you can register in the link below.

RSVP: https://docs.google.com/forms/d/e/1FAIpQLSdK2AxGU9dlgWXinLS6B1jdciH7djGsjeHg3ZOsr3oxOC2QcQ/viewform

All the talk materials will be available here: https://github.com/RamiKrispin/forecasting-at-scale

Thanks to Prof. Steven Ge for the invite! 🙏

ramikrispin, to python
@ramikrispin@mstdn.social avatar

(1/2) Timetk for Python 🚀🚀🚀

The timetk, one of the main R packages for time series analysis and forecasting ❤️, by Matt Dancho, is now available in Python 🐍. The package provides a variety of tools for working with time series data and analyzing it. The Python version leverages pandas for processing time series data and plotly for visualization.

image/png
image/png

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

My R Shinylive app is now on Github Pages. It was simple and straightforward to deploy the app on Pages, a tutorial to follow...

Please be patient while loading 😝

Pros:

  • Serverless
  • Easy to deploy as a website on Github Pages (and similar)

Cons:

  • Slow load time
  • Still experimental

https://ramikrispin.github.io/shinylive-r/

ramikrispin, to datascience
@ramikrispin@mstdn.social avatar

I spent last night to build this fun Shinylive app - Forecasting Sandbox 😎

The app provides a simple sandbox for three simple forecasting models - Linear regression, ARIMA, and Holt-Winters, and it entirely runs on the browser!

I am planning to deploy it to Github Actions and create a tutorial (WIP) 👇🏼

https://github.com/RamiKrispin/shinylive-r

video/mp4

  • All
  • Subscribed
  • Moderated
  • Favorites
  • relationshipadvice
  • rosin
  • Youngstown
  • thenastyranch
  • khanakhh
  • slotface
  • tacticalgear
  • ethstaker
  • kavyap
  • cubers
  • DreamBathrooms
  • rhentai
  • everett
  • magazineikmin
  • bokunoheroacademia
  • InstantRegret
  • Durango
  • osvaldo12
  • cisconetworking
  • normalnudes
  • modclub
  • GTA5RPClips
  • mdbf
  • lostlight
  • HellsKitchen
  • Leos
  • tester
  • sketchdaily
  • All magazines