Statistical Analysis Stack Exchange
Statistical Analysis Stack Exchange

Questions tagged [prophet]

An automatic forecasting system developed by Facebook. Use this tag for any on-topic question that (a) involves Prophet either as a critical part of the question or expected answer and (b) is not just about how to use Prophet.

Prophet is an open-source forecasting method that models trend, seasonality and holiday effects. It is available for both R any Python. The pertinent article is Taylor & Letham (2018), "Forecasting at Scale", The American Statistician, 72, 37-45.

43 questions
15
votes
4 answers

Is Prophet from Facebook any different from a linear regression?

So what I've read about Facebook's prophet is that it basically breaks down the time series into trend and seasonality. For example, an additive model would be written as: $$ y(t) = g(t) + s(t) + h(t) + e_t $$ with $t$ the time $g(t)$ the trend…
K.Hua
  • 263
  • 2
  • 7
13
votes
3 answers

Why is non-iid noise so important to traditional time-series approaches?

I've been reading the whitepaper that accompanied Facebook's release of Prophet, it's time-series modeling library. One topic the authors drew attention to was that noise was assumed to be iid; they note, that this assumption goes against the grain…
jbuddy_13
  • 1,578
  • 3
  • 22
8
votes
1 answer

How does facebook prophet handle missing data?

The Prophet paper (forecasting at scale by SJ Taylor - 2017) says the following on missing data: Unlike ARIMA models, the measurements do not need to be regularly spaced, and we do not need to interpolate missing values e.g. from removing…
Chaymae Ahmed
  • 287
  • 3
  • 7
5
votes
1 answer

Facebook prophet gives a very high MAPE, how can I improve it?

I have some daily sales from 2018-01-01 to 2021-10-21 and I'm trying to predict the sales a year into the future. I opted for facebook prophet. My raw data looks like this: According to a DF-test, the series is stationary. However according to the…
Parseval
  • 295
  • 1
  • 7
5
votes
2 answers

Inference in Time Series: Prophet vs. ARIMA

I read through Prophet's white paper and they mention that their algorithm, "gives up some important inferential advantages of a generative model such as an ARIMA." (page 7) So now I'm curious, what advantages does an ARIMA have for inference? In…
Kylejcaron
  • 53
  • 1
  • 4
3
votes
1 answer

Standardise components of an additive model output

I've got a sales forecasting model using the fbprophet library. The model is additive: calculates a base trend and then adds modifiers for each component/feature, like weekly seasonality, yearly seasonality and temperature. Normal example: sales =…
repoleved
  • 131
  • 5
2
votes
0 answers

How to transform a daily average temperatures forecast into an hourly forecast based on the hourly temperature profile observed historically?

I need to transform a daily average temperatures forecast into an hourly forecast based on the hourly temperature profile observed historically. I work in Python. I have found ways of forecasting the temperatures based on historical values (either…
noemiemich
  • 21
  • 1
2
votes
0 answers

What is the best approach to grouped time-series forecasting?

Let's say we have data on the number of clicks per user over quite a long period of time. We can use, say Facebook Prophet, to forecast daily values given that we have enough historical data. That will be the total number of clicks for all…
Don Draper
  • 86
  • 12
2
votes
1 answer

Adjustment factor in logistic growth model of facebook-prophet

In the paper it emphasizes: "When the rate k is adjusted, the offset parameter m must also be adjusted to connect the endpoints of the segments. The correct adjustment at changepoint j is easily computed as:" $$\gamma_j = \left(s_j - m -…
CheeseBurger
  • 581
  • 1
  • 9
2
votes
0 answers

What are the underlying statistical differences between the vector autoregression model and Prophet?

I am trying to understand the underlying fundamental/statistical differences between vector autoregression models and Facebook's Prophet, with regards to multivariate time series forecasting. I am very new to time series forecasting, but I am really…
Darcey BM
  • 33
  • 5
2
votes
1 answer

Why do time series, with positive values decomposed into seasonal plots, have negative values

I have been running different forecasting algorithms such as Facebook Prophet and Forecast (from R) and I note that despite all my time series values being positive, my seasonality values are negative. For instance, here is the airline passengers…
desert_ranger
  • 245
  • 1
  • 8
2
votes
1 answer

Is summing daily forecasts a sound method for generating weekly/monthly forecasts?

I'm new to time series analysis, and I am wondering if this is a sound method for generating weekly and monthly predictions. In my case, I need to generate daily, weekly, and monthly predictions. If I generate daily predictions for a quarter out,…
smxx
  • 21
  • 1
2
votes
0 answers

How do additional regressors and seasonality get incorporated into fbprophet?

I've been playing around with fbprophet here and I've noticed they specify their base additive model is $$y(t) = g(t) + s(t) + h(t) + \epsilon_t$$ My question is two-fold: 1) How does modifying the seasonality_mode='multiplicative' affect the…
bbss
  • 23
  • 2
2
votes
1 answer

Estimate the time series like an event was never happened

I have data from a website where a specific advertising campaign happened a couple of years ago. What I want to do is to estimate how the signups on that website would have been without that big campaign. In specific, I have the signups of every…
Tasos
  • 93
  • 8
2
votes
3 answers

Forecasting daily data with zeros in Python

I'm currently testing some forecasts on daily sales quantities. However, out of ~2000 observations I have 16 zeros. How should I approach this? It's mainly Sundays and holidays that holds zero as value. I want to perform some transformations to the…
Artem
  • 155
  • 5
1
2 3