Questions tagged [forecasting]
Prediction of the future events. It is a special case of [prediction], in the context of [time-series].
3,878
questions
1
vote
1
answer
406
views
Models that train on Mean Absolute Error or similar?
I'm trying to do time series prediction and I'm interested in training on MAE or other custom loss functions. For my problem I'd prefer having errors of {0, 10, 0, 10, 0, 10} as supposed to {5, 5, 5, ...
0
votes
0
answers
70
views
The time series shows random walk behavior from PACF and ACF plot but adf test shows its stationary
I am new to time series and wanted to practice it by forecasting an hourly time series.
The adf and kpss test results show that the series is stationary.
ADF test results
...
1
vote
2
answers
43
views
Gridsearch on ARIMA favours random walk
I am working on a time-series forecasting problem with ARIMA.
Since long-term predictions were not good, I've started using a "rolling ARIMA" like explained here
...
0
votes
1
answer
1k
views
Are conditional mean in an AR(1)-GARCH(1,1) equal for different GARCH(1,1) processes of the same data?
I have created a Markov-Switching GARCH model, where the volatility is defined to be switching between two different GARCH(1,1) processes. The data is assumed to have zero mean, where the data is ...
1
vote
0
answers
63
views
What's the nowcasting "bible"? [closed]
Is there an accepted best text about nowcasting that you would recommend for someone getting into the field?
16
votes
5
answers
13k
views
How to model prices?
Cross-posted at Math.SE.
I'm working on a hobby project and would need some help with the following problem.
A bit of context
Let's say there is a collection of items with a description of features ...
3
votes
2
answers
670
views
change point detection of time series
Currently, I am working on a research project that involves forecasting electricity consumption using data mining. In my analysis, I have detected the change point for time series using the ...
0
votes
0
answers
38
views
How to get an hourly forecast from mean, max forecast and historicals
I have hourly historical temperature curve for a month say January. I also have a monthly peak and a monthly mean forecast for March 2024 (two values). Using this - How can we get an hourly forecast ...
1
vote
0
answers
37
views
Decomposition time series
I am studying a daily dataset from a game which contains informations about the peak of players from 2013-2023. This is my first try applying decomposition to see how time series components behaves. ...
6
votes
2
answers
826
views
How to incentivise AI to make risky predictions
I'm trying to build a weather forecasting AI. I have a dataset that contains the peak temperatures for each day. I have trained it with MSE as the loss function and it has worked fairly well. I do ...
3
votes
1
answer
482
views
How do quantile time series forecasts work?
My office leadership is interested adopting “quantile time series forecasting”, the idea is query the model to predict the 5th, 25th, 50th, 75th and 95th percentiles of an RV given features such as ...
2
votes
1
answer
85
views
How to implement Dynamic regression Forecasting using only lagged values of the dependent variable
I understand that Regression with Arima errors which is also called Dynamic regression is normally implemented using exogeneous variables, is it possible to implement dynamic regression using only ...
1
vote
1
answer
17
views
How does autoregressive training help limit compounding errors at inference?
I'm having a little trouble justifying something in my head and was hoping someone could provide some intuition? I understand for LSTM models or models that maintain some state about a sequence that ...
4
votes
1
answer
117
views
ARIMA model on yearly data using R
Im new to time series forecasting, and I was trying to model the folowing time series data using ARIMA model in R, so that I can predict for the future 10 time periods.
data: cereal_dataset
When I ...
2
votes
2
answers
453
views
Scoring rules for time series data
I have found quite a lot of articles about scoring rules that seem to first work out theorems and proofs for scoring rules in an iid setting, after which they proceed to apply them to some time series ...
2
votes
1
answer
306
views
Can i use Diebold Mariano test for comparison of 2 models across multiple time series?
I have 2 models (for simplicity, let's call them AR(1) and MA(1)) each making 1 day ahead forecasts of time series.
If I had only 1 time series I would just use ...
5
votes
3
answers
132
views
Quality measure for predictive Highest Density Regions
An alternative to point, interval and density forecasts/predictions would be "predictive highest density regions (pHDRs)", i.e., HDRs for the conditional density of a yet-unknown future ...
0
votes
0
answers
7
views
How to Assess Predictive Potential in Time Series Analysis, Especially with Deep Learning? [duplicate]
In the realm of time series forecasting, how can one assess the predictive potential of a given time series? While traditional methods involve checking for stationarity and white noise characteristics,...
2
votes
1
answer
1k
views
Flat forecast of trended time series data in R
I have a monthly time series of online visits for last 3 years starting from Jan 2016 to Dec 2018 and need to forecast for 2019. The data clearly has an upward trend although no seasonal lags ...
0
votes
0
answers
26
views
Approaches for forecasting higher frequency data with mixed (high and low) freq exogenous variables?
I have y data at a daily frequency and a number of x variables at daily, weekly and monthly intervals. I'm looking to create a ...
1
vote
1
answer
631
views
Breakpoints and Forecasting with R
I am new in R and Timeseries analysis and forecasting. I have 2 questions.
I am detecting three change points in my dataset.
...
15
votes
4
answers
3k
views
Predictive models: statistics can't possibly beat machine learning? [closed]
I am currently following a master program focused on statistics/econometrics. In my master, all students had to do 3 months of research. Last week, all groups had to present their research to the rest ...
1
vote
0
answers
36
views
Valid forms of exploratory data analysis for time series that don't assume stationarity?
Lets say we are given a time series sample and want to try to create a model to forecast future values of said time series
When trying to build a model to forecast time series data, many statistics ...
0
votes
0
answers
36
views
How to use LSTM, TFT, or other RNN with time series of different lengths
I have a dataset of financial transactions per day with the volume of the transaction and the price of the transaction. For each day I want to calculate the volume weighted average price at the end of ...
1
vote
1
answer
53
views
Which machine learning methods that leverage historic and real-time data should be considered for timeseries short-term forecasting?
Some clarifications to my question:
The data I have available for use is: (a) historic data of features and ground truth on 60-minute interval, (b) real-time data of features on 60-minute interval, (...
0
votes
0
answers
84
views
Avoiding negative values in Forecasting
I have questions about the Box-Cox transformation that can be used to maintain positive forecasts (Log-Transform) when $\lambda=0$
\begin{equation}
w_t =
\begin{cases}
\log(y_t) & \...
3
votes
1
answer
1k
views
Forecasting with AR(1) and pseudo out-of-sample using R
I'm trying to do Pseudo out-of-sample forecasting using R. And, I also have the following initial data (gdp)
...
3
votes
1
answer
429
views
If extrapolation is bad then how is forecasting methods statistically relevant
There are lot of articles out there that talks about why extrapolation is a bad thing to do.
My question is if the above is true , how are forecasting methods like forecasting the trend based on some ...
1
vote
1
answer
244
views
Should bootstrapped prediction intervals be normally distributed?
I am trying to implement boostrap prediction interval example of FPP3 book in python for learning purposes (https://otexts.com/fpp3/prediction-intervals.html).
Prediction interval is estimated by ...
2
votes
1
answer
995
views
Regression with ARIMA Errors for non-stationary timeseries: Mixing of stationary/non-stationary covariates?
Given I want to forecast e.g. monthly sales (dependent variable, likely non-stationary) with regression and ARIMA errors (ARIMA in R with xreg) I have e.g. two independent variables/covariates:
...
1
vote
1
answer
148
views
Time Series vs. Queueing Models
Generally speaking, queues are modelled using the Poisson process. Supposedly, this used to model the dynamic nature of queues, arrivals, birth-death and renewals.
But just as a basic question: Why ...
3
votes
1
answer
135
views
Forecasting based on revenue and advertisement
How to forecast revenue based on previous revenue and advertisement budget?
This is how the data looks like, see full data on google sheet or even better on Colab notebook.
EDIT:
With my limited ...
3
votes
1
answer
3k
views
How can I forecast with multiple time series sampled at different frequencies?
I am attempting to build a time series model that can predict order volumes for a single firm in the trucking industry.
I have time series data on the firm's order volumes. This data is monthly data.
...
0
votes
1
answer
2k
views
Forecasting with panel data/ time series
I have three questions for you regarding the prediction of panel data.
There is a function predict.plm() in the plm package for <...
2
votes
1
answer
952
views
Does seasonal differencing in SARIMA model take care of additive/ multiplicative seasonality?
I am exploring the use of ARIMA and Seasonal ARIMA models (SARIMA). In some of my datasets, I can clearly observe seasonality in the ACF and PACF plots (the lines at seasonal lags clearly cutting the ...
4
votes
1
answer
88
views
Very narrow confidence interval in ACF and PACF graphs
I am trying to forecast electricity consumption using hourly consumption data over a span of 4 years. However, when I plot the ACF and PACF graphs, the confidence interval turns out to be very narrow, ...
1
vote
2
answers
82
views
Forecasting Methods for a Short Time Series with No Trend or Seasonality in Python
I am pretty new into data science and I had some issues with my project.
I am trying to build a forecasting model for a time series data. It is about yearly CO2 emissions from agriculture.
The issue ...
2
votes
2
answers
493
views
Predict number of users
I have historical data (daily, weekly, monthly, however I want to slice) for a few years and I want to predict the probability of hitting an end of month target throughout the month. The data follows ...
0
votes
1
answer
375
views
How to deal with Covid outlier in time series/machine learning forecasting?
Disclaimer: I checked some similar questions but I could not find anything in particular that would work for my case.
I am dealing with a time series going from 2015 to 2023. The data points are the ...
4
votes
1
answer
2k
views
Adding noise to time series data to increase training data
I am dealing with a weekly time series forecasting problem and I am currently investigating the use of an LSTM to make a multi-step forecast for a univariate time series. I actually have a ...
0
votes
1
answer
850
views
Determining order of ARIMA(p,d,q) from ACF and PACF
I know that when trying to determine if you have an AR(p) or MA(q) process, you look at the PACF and if it drops off significantly at a lag p, then you can say it's an AR(p), but if it's geometrically ...
1
vote
1
answer
35
views
When I am comparing RMSFE between a log model and a level model of the same dataset, how should I proceed?
I have created an AR(2) and AR(5) model to forecast my data in stata. The AR(2) model is a level model, while the AR(5) model is a log model. I have computed the RMSFE to compare the forecasting ...
3
votes
1
answer
166
views
Does the absolute value of MAPE or sMAPE have meanings?
Let say I have a forecasting system compared with a naive forecast that just use the today's value as forecast. If the naive forecast have a MAPE of 200%, and my system have a MPAE of 100%, could I ...
2
votes
1
answer
74
views
EDA and Model Selection for Forecasting while avoiding Data Leakage
How to do EDA and model selection for time series forecasting without data leakage?
Im assuming just checking for missing values is ok. But is graphing the entire time series considered data leakage?
...
0
votes
0
answers
35
views
Forecasting cash inflow using network load usage - timeseries vs GAM (or similar)
I have two datasets of daily cash_inflows & network consumption (load) for past 4 years. Network load consumption is the single source of income. There will be some lag between load consumption &...
0
votes
0
answers
42
views
How can I obtain the distribution of the number of customers in a non-starionary MM1 queue given an interval partition of stationary transitions?
I am considering this transient solution for the probability mass function over the number of "customers" in an MM1 queue:
$$ p(k; t, i, \lambda, \mu) =\\ \exp \left( - (\lambda + \mu) t \...
0
votes
0
answers
45
views
Is there a heavy traffic approximation for percent under benchmark?
Suppose I have an waitlist of patients waiting to be served. Each patient has a benchmark number of days that the service should be completed by as a goal (not a hard constraint of the modelling). ...
2
votes
2
answers
955
views
Rolling Window Forecasting with ARIMAX while supplying actual values
I am comparing different exogenous variables in how good they support the forecast of the monthly seasonal adjusted unemployment rate. All my data is monthly (2006-01-01 until 2018-09-01) and ...
1
vote
0
answers
20
views
Is there a known way of producing forecasts with reasonable fit and residuals that are at least independent, & ideally negatively correlated? [closed]
I am trying to do some forecasts. I have produced multiple forecasts by a variety of methods. All of the forecasts I have generated so far have residuals that are strongly positively correlated.I ...
0
votes
1
answer
24
views
Statistical or Machine Learning Approach for Simulating Variable (proportion)
I am currently facing a challenge in analyzing customer satisfaction data from an electric drive production facility. During a specific period of the month, we encountered technical issues with our ...