Questions tagged [forecasting]
Prediction of the future events. It is a special case of [prediction], in the context of [time-series].
3,222
questions
1
vote
0answers
16 views
Root-mean-square error when having multiple prediction horizons
I have a basic question about the root-mean-square error (RMSE). I have a prediction using an ARIMA model. I predicted a time series and use a rolling-horizon approach with overlapping or non-...
1
vote
2answers
21 views
consistent estimation of quantiles (without overlapping quantiles)
I would like to forecast quantile ranges. The observations are assumed to be heteroscedastic. Mostly, I am confronted with the problem that quantile regression results for different quantiles do ...
2
votes
1answer
32 views
Should the day of the week variable be splitted into 7 columns or is one enough? Time-series forecast
I have time-series data, in which I added a variable "dayofweek" with the day of the week varying from $0$ (Monday) to $6$ (Sunday) (Python's default). I'm using boosting models like GBM and ...
1
vote
0answers
17 views
Confusion on optimal lagged time and other associated parameters to use for attrition forecasting
I am working on employee attrition prediction with raw data that has more than 1 year's amounts of daily metrics, and have some confusion on the best data aggregation techniques/methods for a ...
0
votes
0answers
22 views
Generic time series model not bound to a specific series
I have a large set of relatively simple time series with very similar behaviour, on which I would like to do short-term forecasts. These series are non-aligned, and at one moment in time, only a ...
2
votes
1answer
33 views
One step-ahead forecasts in R
I would like to compare one-step ahead forecasts on a given time series for ARIMA and UCM (using KFAS library). I have split my time series in train and validation, that I will use to understand which ...
0
votes
1answer
20 views
How can I forecast Gini using ML?
I have a data set containing 20 years of Gini values for a country. The latest data are for 2018. I want to predict the Gini values for this country by 2025. How can I do this using ML techniques? ...
1
vote
1answer
27 views
Can I measure the accuracy of a range of quantiles in my forecast distribution?
I am forecasting items and measuring the point forecast and distribution accuracy of numerous different models against actuals. To measure distribution accuracy I am using the continuous-ranked ...
0
votes
0answers
8 views
EViews: VAR dynamic forecasting data
In Eviews, I tried VAR out-of-sample dynamic forecast.
But after doing forecast process, there are all same data left.
For example, I used 1998q1~2008q4 GDP growth rate, exchange rate, etc...
to ...
1
vote
0answers
15 views
Forecasting with external regressors: correlation and causality between the variables
I have to forecast the value-added growth rate of the Italian service sector, and I would like to do that with an ARIMA model with an external regressor. By looking at the dataset at my disposal and ...
0
votes
0answers
22 views
Understanding Facebook's Prophet
I am having trouble with understanding facebook's prophet's cross validation. I have an ARIMA model that splits the data into training and test sets and performs a rolling forecast by using the train ...
1
vote
1answer
22 views
Can a Simple ANN be Generative?
If a simple ANN was trained to predict the next step in a sequence, such as a univariate time series, can it be considered a generative model?
2
votes
1answer
28 views
Forecasting multiple hours with lag variable - XGBoost or GBM
I have an hourly time series data like this:
...
0
votes
0answers
17 views
Forecast automation when dealing with many time series
When working with a time series, it's worth plotting it and analyzing its features, such as trend, seasonality, etc. These features suggests some models for you to try out. In the end, you choose the ...
1
vote
0answers
20 views
Is it possible to calculate AIC on LSTM?
I'm doing a forecast on returns of stocks using ARMA-GARCH models and LSTM. Because of the nature of the data, RMSE, MSE cannot be used. I instead found MAAPE.
Now what I'm trying to understand if I ...
1
vote
1answer
25 views
Do tree methods like gradient boosting predict all iterations at once?
If I'm using a tree method (e.g GBM) and I have a time series hourly data, and I predict my target variable $y$ for the next 48 hours, do my predictions were made all at once, or does the second day ...
0
votes
0answers
19 views
Forecasting new product sales - pre launch
Product X is due to launch in 6 months time and I am looking for a method to estimate what the sales of this product could likely be in 3 years time.
Product X belongs to a category of products for ...
0
votes
0answers
153 views
ARIMA models for vote percentage time series
I am currently a first-year analytics student, so far I have understand that ARIMA models are suitable to be used for time series modelling, however, they are not suitable to fit time series like e.g. ...
0
votes
0answers
24 views
Can I use Granger causality for missing data/ irregular time series?
I have two time series and I want to look at their Granger causality. My data is from two irregular time series. Irregular here means sampling was done sporadically across 3 years. In some months, ...
0
votes
1answer
67 views
Auto forecasting vs hierarchical forecasting
I'am a new data scientist in my company, and we don't have expert to guide me, so please i really need for your help and guidness for this problem.
I need to forecast sales for many products and i ...
0
votes
0answers
29 views
Calculate optimal linear forecast based on infinte past?
Consider the following $MA(1)$ process : $$X_t=\epsilon_t-\theta.\epsilon_{t-1}$$
Where $(\epsilon_t)_{t\in \mathbb{Z}}$ is a white noise whose variance equals $\sigma^2$.
How do I calculate the ...
0
votes
0answers
13 views
How to do Multi-step time series prediction in spss modeler python scripting
The idea is to start with an initial prediction for a single time-step. After that then I add the predicted value to the input values for another prediction, and so on. In this way, I will create ...
0
votes
1answer
31 views
How to calculate the RMSPE when the data contains zeros?
I made some predictions with my model and measured its performance using the RMSE. I cannot compare the errors between predictions because some time series have different scales. Some in the 1000's ...
0
votes
0answers
16 views
Can two sources of information be combined to get a better prediction than given by either multiple regression or time series analysis?
Suppose I have a data set, which I will call Height-Weight-Age featuring annual height and weight measurements of $N$ people from ages 10 to 30, where $N$ is relatively large, e.g. $N \approx 5000$. I ...
2
votes
2answers
111 views
Predicting millions of independent time series, using them to help each other
This is a very general problem faced by different types of companies. Predict future customer behavior over time.
Imagine that we have 1 million customers with their own resources over time, forming a ...
1
vote
1answer
33 views
Unreasonable bias when using nnet (R package caret) for time series forecasting
I have been trying to forecast a time series in a regression-like setting using neural networks (nnet method in R package caret)....
0
votes
0answers
18 views
Forecast combination with nonstationary time series
Suppose I have a non-stationary time series and have obtained forecasts using various methods such as ARIMA, ETS, Theta etc. I want to find a weighted combination of these forecasts.
I found in the ...
0
votes
1answer
22 views
R: Density Forecasts using rugarch
I would like to construct a density forecast using a GARCH model. Is it possible to use rugarch in R to construct these? For example, using a ARMA(1,1)-GARCH(1,1) ...
1
vote
0answers
28 views
Why research on time series always use data with the same (instead of lower) level of precision to make forecast?
Suppose there is a dataset with data points for every 1 second:
x1, x2, ..., x15
and suppose the following are the average values for every 5 seconds:
...
1
vote
1answer
36 views
ARIMA fit vs. Linear Regression — which one to use?
I have a data set made up of 30 observations, years 1980-2020. The second variable is the amount of cargo passing through a particular harbor on the west coast (in tons). I have attempted to fit an ...
2
votes
1answer
38 views
MAD/Mean ratio disadvantages?
I've been trying to assess the best accuracy measure, especially for intermittent demand. While I have found scaled measures such as RMSSE and MASE to be really good measure I find that their ...
0
votes
0answers
18 views
How using a Fourier Series to model seasonal data with ARIMA works?
My understanding is that if we have a seasonal time series with period of $m$ we could model seasonal data by approximating the time series as $y_{t} = a + \sum\limits_{k=1}^{K}[\alpha_{k}]\sin(2 \pi ...
0
votes
0answers
28 views
Predict a Time Series With a Multi-Level Categorical Independent Variable
I need to perform and evaluate predictions on a time series with some 3,000 categories as independent variable. Specifically, I have 3 fields:
"Value", which is my independent variable (the ...
0
votes
0answers
35 views
Difficulties and possible adjustment measures of forecasting stock prices
I understand that a non-stationary time series have fluctuating mean and variance (and ACF) over time. In my case for stock prices, it is leaning towards having cyclic pattern rather than seasonal ...
0
votes
0answers
16 views
Forecasting with Dynamical Bayesian Networks
I am trying to forecast some variables of a dataset (time series) with Dynamical Bayesian Networks (DBN) using pgmpy. I could be mistaken, but what is being called "forecasting" in the ...
1
vote
0answers
13 views
R Arima model has negative variance (sigma2) [closed]
I am trying to fit an arima model on my dataframe with monthly data, with both fable::ARIMA and forecast::Arima functions.
Here ...
0
votes
1answer
43 views
What does the integrated part of the Arima model do?
I understand that the integration is used to 'convert' a non-stationary process into a stationary process using the method of differencing. I'm just unsure exactly what differencing by $d > 1$ ...
0
votes
0answers
10 views
Decomposition plot interpretation
I have a question about the interpretation of a decomposition plot. From Hyndman's book "Forecasting: Principles and Practice"
which results from a classical additive decomposition model of ...
0
votes
0answers
9 views
Is it possible to adjust the estimated parameters in VARMA model to satisfy given moments in the forecast?
Does anyone know a way adjust the parameters in the VARMA model so that the h-step forecast satisfy some given expected values, standard deviations and correlations while changing the model as little ...
1
vote
1answer
13 views
Interpretation of scaled error measures
can someone give me an explanation on how one would interpret the result of a scaled error measure.
For example the Mean Absolute Scaled Measure (MASE). The numerator is the mean absolute error and ...
-5
votes
1answer
73 views
Multi-variate time series forecasting with Python [closed]
I have the following min-max normalized 2D dataset:
The actual dataset has 32 features + y, and several thousand instances (features 1-5 start having different values further down the line).
I plan ...
0
votes
0answers
17 views
Sales forecasting
I'm trying to do a sales forecast of an existing product in a new market.
I've been doing research and I think new product sales forecasting is the right characterization of the problem. Am I right to ...
0
votes
0answers
12 views
Gap between original data and forecasting time series
I forecasted a set of data, 'ar1.s' from the TSA package, it was meant to be an AR(1) process.
The code I used is:
...
0
votes
0answers
11 views
Dynamic Regression vs linear regression oddity
I tried a regression using tslm in R of the "medicaid participation rate" on trend, poverty rate lagged, and a dummy indicating after 2014. The fit is ...
0
votes
0answers
16 views
Choosing proper Frequency argument for time series object
I have a year-long data of an anemometric tower. The variable that I'm interested in is wind speed at 50m. My idea is to train with this data using some models (Arima, Sarima, Holt-Winters, NNETAR, ...
0
votes
0answers
13 views
Auto.arima application in R with sub-daily data - No seasonality given [duplicate]
I'm working on a R code and my aim is to make forecasts with a model chosen by applying auto.arima function on my data. These are recorded every 6 minutes, so we're dealing with sub-daily data.
...
2
votes
1answer
19 views
What is this measure of error called?
I am investigating prediction errors in a context where the errors can be extremely large. Someone advised me that in addition to reporting the mean or median of the absolute errors $$|\hat{x} - x|$$ ...
0
votes
0answers
16 views
Multivariate Time series forecasting- Statistical methods
I was trying to forecast the truck numbers required at each distribution location...for that I was forecasting the shipments(number of units) at each location and dividing it by a factor to get the ...
0
votes
0answers
6 views
Analyze source of forecast error
I have inventory forecasts and inventory actuals for every month for the previous year. Results are around 2% forecast error and I want to know the parameters that cause it (I have many parameters ...
1
vote
1answer
14 views
Ambiguity in definition of Long Term Forecasting
The well known definition of a long term forecasting model is based on the length of the forecast horizon. In case of long term, it is usually agreed to be in the order of years (ahead of the most ...
