Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Prophet: forecasting at scale

research.fb.com

Post reply on HN
Print view View original

71–80 of 114 posts

Re: Prophet: forecasting at scale

#71

Post by T-A »

etjossem wrote: post #33
blazespin wrote: post #3

Earlier quoted context omitted.

Probably just help verify that the stock market is a random walk with a meager trend upwards that doesn't beat inflation + trading costs.

> Probably just help verify that the stock market is a random walk with a meager trend upwards that doesn't beat inflation + trading costs. That doesn't sound right. Let me clear that up for you. Since 1950: S&P 500 Annual Price Change: 7.2% S&P 500 Annual Div Dist: 3.6% S&P 500 Annual Total Return: 11.0% Annual Inflation: 3.8% Annual Real Price Change: 3.3% Annual Real Total Return: 7.0 % Buying the straight S&P 500…

Buying the S&P 500 in 1950 and holding 67 years does.

One sample tells you nothing about randomness. What if you buy in August 1929? What if you hold for a more realistic 20 or 30 years from peak earning years to retirement?

Re: Prophet: forecasting at scale

#72

Post by feld »

dmichulke wrote: post #64

I have been working for a few years on a similar project using evolutionary algorithms on top of other models (linear / ann). It works quite well (e.g., for equidistant energy demand / supply forecasts) but there's still lots of stuff to do. It's major benefit is that it figures out relationship to the target time series by itself, so you can just throw in all time series and see what comes out. Language is Clojure,…

Why not release sooner and document the code smells? Maybe you'll get patches

Re: Prophet: forecasting at scale

#73

Post by JoelSanchez »

Cthulhu_ wrote: post #67
JoelSanchez wrote: post #47

Earlier quoted context omitted.

https://tools.wmflabs.org/pageviews/?project=en.wikipedia.or... What's up with Java? (Set "logarithmic scale" to improve the visualization)

the wiki page is about Java the country, not the programming language. Haven't found any relevant news around that time though.

Indeed, that's why I used "Java (programming language)".

Re: Prophet: forecasting at scale

#74

Post by unoti »

techno_modus wrote: post #35

It seems that they have developed a model for only univariate forecasts and only numeric regular time series which is a classical use case in statistics. Yet, most data sources have many dimensions (for example, energy consumption, temperature, humidity etc.) as well as categorical data like current state (On, Off). The situation is even more difficult if the data is not a regular time series but is more like asynchr…

> most data sources have many dimensions (for example, energy consumption, temperature, humidity etc.) as well as categorical data like current state (On, Off). The situation is even more difficult if the data is not a regular time series but is more like asynchronous event stream. It would be interesting to find a good forecasting model for some of these use cases.

I'm guessing you already know about this based on the way you described the situation, but the Hyndman Forecasting book [1] discusses various models at length for doing multivariate forecasting models. It's loaded with code and samples in R.

1. https://www.otexts.org/fpp

Re: Prophet: forecasting at scale

#75

Post by vinw »

T-A wrote: post #69
vinw wrote: post #62

Earlier quoted context omitted.

FIS Prophet is targeted at actuaries, and really no-one else so I don't know if anyone will care. They have had the name a lot longer than Facebook though!

This other Prophet has also been around for a while: https://github.com/Emsu/prophet

Right, but some of the source code in FIS/Prophet goes back to the 1980s.

Re: Prophet: forecasting at scale

#77

Post by mej10 »

dragandj wrote: post #37
confounded wrote: post #11

Worth noting Prophet is R/Python wrappers to some models with reasonable defaults, written in and fit by Stan , a probabilistic programming language, and Bayesian estimation framework. Stan is amazing in that you can fit pretty much any model you can describe in an equation (given enough time and compute, of course)! More on Stan here: http://mc-stan.org/

... and if you like Clojure, you might try Bayadera, which has its own engine running the analysis on the GPU. http://github.com/uncomplicate/bayadera

This looks like it could be awesome but it has almost no information about what its purpose is or how to use it.
Post reply on HN