Questions tagged [forecasting]

Prediction of the future events. It is a special case of [prediction], in the context of [time-series].

Filter by
Sorted by
Tagged with
1 vote
1 answer
11 views

Why when calculating RMSSE (from M5) is the denominator based on training data?

The RMSSE formula from the M5 competition is the following: https://mofc.unic.ac.cy/m5-competition/ This indicates the denominator, which is the naive error, is based on the 'training' data. Below is ...
  • 115
1 vote
0 answers
24 views

Which models to use when forecasting time series data that shows exponential decay?

I'm working through "Forecasting: Principles and Practice (3rd edition)" by Rob J Hyndman and George Athanasopoulos to better understand times series forecasting in an R environment. This ...
0 votes
0 answers
14 views

Leap year in daily time series data

I need to forecast daily electricity demand. There are two leap years in my dataset. I am just allowed to use the forecast library. Is it possible to exclude these two days? I read this but I could ...
  • 267
0 votes
0 answers
22 views

Derivation of variance of ar(1) forecast residual formula

I've included an image of the formula. I'd appreciate it if anyone could provide me with the derivation for it or if they could provide me with a link to anywhere online or a textbook that covers it. ...
0 votes
0 answers
26 views

How to derive a probability distribution from the R fable package forecast function?

In running the R code posted at the bottom, I derive a forecast for the next 10 periods at the 80% and 95% confidence levels, using the fable package and running 1000 simulation sample paths, as ...
3 votes
0 answers
39 views

Do I need to have studied econometrics to do time series analysis? [closed]

I have an undergrad degree in Economics and Management but my academic training/background in econometrics is insufficient, although I did study micro/macro economics and fundamental math and ...
  • 31
0 votes
0 answers
19 views

Time Series: Predict Regressor or Omit Regressor?

Assume a simple time-series Y with no regressors sampled by hour and a time-series N for the same period, also with no ...
0 votes
0 answers
6 views

Does all disturbance models have the static gain equal as 1?

Disturbance signals for a dynamical system is very difficult to measure because they appear everywhere. The disturbance signals is always a normal distributed with zero mean, e.g gaussian signal, ...
  • 193
4 votes
1 answer
50 views

Penalising Error above a certain Threshold

I have a ML model (a NN in the specific but I don't think it's important for the purpose of my question) that is doing pretty decent at his job, which is predicting the demand of a certain substance X ...
0 votes
1 answer
29 views

Why is the prediction interval going way beyond the $100\%$ threshold? [closed]

Prediction interval in my forecasting is too high. It goes beyond the threshold of 100% of forecasting share of health spending as a percentage of total. This is the relevant data and the associated ...
0 votes
0 answers
34 views

What is the best way to normalize a timeserie with a trend without differencing it?

On a multivariate forecasting problem (a target and some covariates with known history used to predict future of the target) i'm struggeling with the normalization of my data (covariates and target). ...
0 votes
0 answers
41 views

How to carry out the recursive estimation of a covariance matrix?

This questions is inspired by my reading of "From Probabilistic Forecasts to Statistical Scenarios of Short-term Wind Power Production" by Pinson et al. (2009). Let $X_k \sim \mathcal{N}(0,1)...
  • 610
1 vote
0 answers
16 views

Correlation of forecast target and forecast error in time-series framework

Setup: In a time-series context, let $y_t$ denote a target variable that is $\mathcal{F}_t$ measurable, and let $h_t$ denote a forecast of $y_t$ that is $\mathcal{F}_{t-1}$ measurable. That is, the ...
1 vote
0 answers
39 views

Recursive Maximum Likelihood Estimation algorithm - The same as Maximum Likelihood Estimation?

I have the book "Adaptive Control", Second edition, from Karl-Johan Åström and at page 61 to 62 he wrote: Stochastic models The least-squares estimate is biased when it used on data ...
  • 193
1 vote
1 answer
25 views

Time series forecasting for dataset spanning a week

I am a newbie in time series data forecasting. I have a week long data and the counts represent arrivals per 5 mins period. A part of the dataset is shown below. ...
  • 11
0 votes
0 answers
12 views

Improving performance of ARIMA with regressors in forecasting

I have some time series data, for revenue, and 4 separate media channel spends data. These go weekly from 2019 to 2023. I also have future media channel spends, for the next 25 weeks. I want to be ...
  • 113
0 votes
0 answers
27 views

Multi-variate multistep time series Forecasting for non-stationary data

The problem is that I have a very special time series. It is sensors data for a machine. I have about 400 sensor data which I want to use to forecast the machine advancing speed. The data contain a ...
  • 1
0 votes
0 answers
21 views

How good are inflation expectations as a predictor on inflation?

I want to research inflation expectations and how well they predict inflation. I have found some past articles but none of them explain how to forecast inflation using inflation expectations. I have a ...
  • 1
0 votes
0 answers
8 views

Convergent and Divergent Prediction Intervals in Time series models

My question is similar to this one Whether increasing the sample size influences the prediction interval? and it's also related with Rob Hyndman - Forecasting: Principles and Practice (2nd Edition), ...
1 vote
0 answers
24 views

How to chain separate models into a single unified model (where predictions of one sub-model form an input to another)?

Our team is tasked with forecasting several timeseries at the daily (or hourly) level: number of calls ('demand') number of calls catergorised as important mean vehical travel time to caller mean ...
  • 121
0 votes
0 answers
13 views

Is there a way to do Dynamic Harmonic Regression in R using multiple variables?

The company I work for would like to forecast weekly transactions, given a certain weekly sales budget (i.e. predicted weekly sales) for a period of time. We are a highly seasonal business, ...
0 votes
0 answers
35 views

Quantity Prediction using past sales data python

I'm trying to predict quantity that can be recommended to store based on the their past purchases. The data used consist of store, product details and respective quantity purchased for one year time ...
0 votes
0 answers
14 views

IRF function in VAR model looks incorrect

Unfortunalty I did not find a answer to my specific question, so i count for your help. Please have a look at my response plot I used the following code: ...
2 votes
2 answers
75 views

Forecasting revenue - what and how to pass input

I have a dataset with quarter wise revenue for past 3 years from Jan 2020 to Dec 2022. I have 4642 customers. Each customer has 1 row of data which includes features based on his purchase frequency, ...
  • 2,290
0 votes
1 answer
40 views

Confusion between the meaning of seasonality and seasonal patterns in time-series forecasting

According to Forecasting: Principles and Practice Seasonality is always of a fixed and known frequency. If it is a fixed and known frequency, does that mean every series with monthly or quarterly ...
0 votes
0 answers
15 views

When is it appropriate to forecast data which were transformed with the log(diff) function?

For a work, we have to apply a Box-Jenkins approach to a certain data. We choose to study the total industrial production in Belgium (monthly data). As we have to forecast the data, we did multiple ...
0 votes
0 answers
10 views

Segregate/Decompose a prediction for a time period into smaller sub-periods

I have a dataset of Electric vehicle demand every 5 mins at every station in a cluster. However, this data is sparse so I cannot train a model and extract the underlying patterns. Therefore, I do some ...
2 votes
0 answers
24 views

How to predict price with dummy variables ARIMAX in R?

Can anyone help? I want to predict a price variable using five dummy variables. Data conditions: contains an up-trend component not stationary I'm confused about whether to split the data or not. It'...
  • 21
0 votes
0 answers
5 views

Continue forecast from the same base while forecasting panel data

I know I'm asking a very general question. I try to forecast on panel data like this; customer time value A 1 10 A 2 20 A 3 25 B 1 100 B 2 110 C 1 500 C 2 525 C 3 600 C 4 400 The real data ...
1 vote
1 answer
40 views

How to estimate a ARCH/GARCH model since we don't know last period conditional variance?

I'm wondering how it's possible to forecast a conditional variance using a GARCH model since we don't know last period conditional variance. If I understand correctly, the conditional variance is the ...
1 vote
0 answers
31 views

Why LSTM predicts very poorly with loss curves showing no indication of overfitting or underfitting?

I am training a LSTM model for load demand prediction with: Training Data: 18288 samples with 9 features Validation Data: 0.05% of Training Data (about 915 samples) Data Scaling: MinMax Scaler(0,1)---&...
0 votes
0 answers
29 views

Recursive time series forecasting test set

Problem: I am building a multivariate model for recursive time series forecasting, where the goal is to make a 4-step-ahead forecast. The actual data for the forecasting period is available. As far as ...
0 votes
0 answers
10 views

How to interpretate Net Difference Raster

I know english till certain level and also tryied translator, but direct translation didn't made it clear enough, then, I need some help to interpretate what is a Net Difference Raster. I'm exploring ...
0 votes
0 answers
29 views

How to compute loss on GARCH multi-horizon predictions vs realized time series

Suppose I have a daily financial timeseries of zero-mean returns r(t), and I fit a GARCH model. I'm using the arch_model package which forecasts volatility by ...
1 vote
1 answer
82 views

What is the correct way to average random variables and get correct quantiles

Say I have two random variables $A$ and $B$ which may or may not be independent. I also have their $0.95$ quantiles $Q95_A$ and $Q95_B$. What is a valid way to average these densities and obtain valid ...
  • 1,691
1 vote
1 answer
37 views

Is Sarima(0,0,0)(5,0,0,12) the same as the 5 year avg of a given month? [closed]

Trying to understand big P,D,Q in sarima better. Could probably ask a broader question to get more info but my first question is making sure I understand at least the seasonal AR term. With m=12 (...
  • 121
1 vote
0 answers
19 views

Good books on STLF (Short-term Load Forecast)

May I get some of your recommendations on this subset of the more general time series forecasting class of problems? The reason I need your help is that there's a big number of books that do not ...
3 votes
1 answer
37 views

Can regression forecasts of univariate time series be independent (of one another)

Suppose I have short-term forecasts from two univarite regression models of the same time series. I am choosing the models to be as different as possible in structure and assumptions. For instance, ...
  • 2,747
0 votes
0 answers
46 views

Arima and Improving Prediction

I have some time series data for online sales, which are weekly and go from December 2014 up until December 2022. My aim is to build a model where I can forecast the revenues for the upcoming weeks. ...
  • 113
0 votes
0 answers
19 views

AR Modelling. Box Cox and Differencing don't give Stationary Data

I am trying to fit an AR/ARIMA model on electricity data on hourly prices during a almost three year period, I am following the guidelines in Hyndman, R.J., & Athanasopoulos, G. (2018) to do this. ...
0 votes
0 answers
25 views

Do I need to constantly incorporate data to my forecasting model?

I'm new to the forecasting world and I have a very basic question to ask. English isn't my first language so I apologize in advance for any mistakes in my grammar. I'm trying to forecast daily data. I ...
0 votes
0 answers
22 views

state space models implementation in R

I'm trying to implement a state space model in R with daily data. Is there any package to do this in R? I haven't been able to find one. A little bit of context: I need a model for forecasting gas ...
1 vote
1 answer
26 views

Force predictions of 2 time series models with different steps to be consistent

Suppose I have a time series. Let's say it is of the number of sales in a shop. Suppose I am looking to make two models - model 1 which predicts future values by weekly time steps (total sales per ...
  • 11
1 vote
2 answers
52 views

Short-term gas demand forecasting

I'm a 22 year old Statistics student with a big problem to solve. English isn't my first language, so I apologize in advance for any mistakes in my grammar. I'm trying to make a short-term gas ...
0 votes
1 answer
73 views

Combine probabilistic forecasts with weighting

Suppose I want to compute the probability that Argentinia wins the worldcup semifinal Argentinia - Croatia. I have two independent sources of information about this probability, whereby source A says ...
  • 253
0 votes
0 answers
56 views

Forecasts are at different levels using HTSRegressor - hierarchical time series

I'm trying to forecast a hierarchical time series and while predicting, my forecasts are at random date levels instead of the same level as my training set. How do I handle this. It might be a simple ...
0 votes
0 answers
17 views

How can Linear Regression be used for forecasting? Some features are not available at present time

Given the feature matrix $X\in R^{N\times C}$ and observed prediction $y\in R^{N\times 1} $, an OLS computes the coefficients $\beta\in R^{C\times 1}$ so that $$ \hat y=X\beta $$Now, forecasting asks ...
0 votes
0 answers
22 views

How to find base pricing and how variables effect final pricing?

I have a large set of sales data. I have the Skus and final sales price and a lot of extra information such as date, location, buyer, etc. I am wanting to make a sort of calculator so someone could ...
  • 1
2 votes
1 answer
45 views

Prediction intervals for simple/baseline forecasts

I'm reading this, and I don't understand how the prediction intervals are calculated for the baseline forecast methods. I agree that we can estimate the SD of the sample from the known residuals $$ \...
2 votes
1 answer
49 views

Explaining ARIMA forecasts

Currently attempting to interpret the results of my forecast using an ARIMA model that was applied to time series data (Dataset below). The forecast attempted is for a year into the future. The data ...

1
2 3 4 5
75