Questions tagged [forecasting]
Prediction of the future events. It is a special case of [prediction], in the context of [time-series].
3,579
questions
2
votes
1
answer
58
views
End of a time series have to be "corrected" before decomposition if partial period?
I have a few times series that are Google search terms which a period of length 1 year with 12 months in each year. As we are in February, the latest data shows only one and a half months.
All my ...
0
votes
0
answers
21
views
Coding independent variable for regression based on relationship with dependent variable and reducing dimensionality
I have a table where rows are dates and columns are values of a dependent and many independent variables.
I want to create regression with a number of nominal and ordinal independent variables to ...
0
votes
1
answer
41
views
Forecasting Prices with interdependence that form a Timeseires
I have already asked a simmilar question, but i thoguth that this was not phrased well and hence i am trying a new post were i ask a better question. Let me know if this is ok. Judging by some of the ...
0
votes
0
answers
29
views
Difference between ML models and conventional time series model such as VAR in multivariate time series forecasting
I am learning to use ML model to do the time series forecasting / prediction in multivariate. But I have confusion about them. For example, I have a monthly dataset UCI AirQuality. I would like to ...
0
votes
0
answers
21
views
Why is my AR forecast better than my ARIMA forecast even though data is I(1)
I am trying to build a time series model for forecasting. The time plot of the data is as shown
Evidently, there is a trend component and the series is not stationary. As the interest is in ...
0
votes
1
answer
39
views
Is this an expected behavior of an ARIMA(0,0,4,1,1,1)
I have data of daily sales and want to make a forecast in SPSS modeler, however I'm getting a result that I can't explain. The plot is inserted below. The fit is the red line and the blue is the ...
1
vote
0
answers
17
views
How to deal with the miss# how to deal with the missing values in time series forecasting?
I have dataset in which there are records of stock market hour by hour. There are some missing values like each day should posses 24 values as it is hour
by hour but I don't have the value of some ...
0
votes
1
answer
54
views
Final model for time series forecasting
I am struggling with understanding how the final performance on an independent test set is obtained in a time series forecasting scenario.
My understanding is that rolling CV is used (as seen as in ...
2
votes
1
answer
130
views
How does the ARMA forecasting method using the recursion equation and replacing past errors with residuals make sense?
The forecasting method described here doesn't make sense to me.
Suppose we have a MA(1) process $Y_t = \epsilon_t + \theta \epsilon_{t-1}$. Let $\hat{Y}_t$ be the forecast of $Y_t$ at time $t-1$. We ...
1
vote
1
answer
39
views
Negative forecast value using ARIMA from fable and NaNs warning using ARIMA
My data set is a weekly data that contains two variables Production and Shipment. Production is the independent variable and Shipment is the dependent variable. First I'm trying to forecast Production ...
0
votes
1
answer
82
views
Time series forecasting: classical methods vs. machine learning / artificial intelligence [closed]
Do you know papers that researched how accurate predictions are using time series or AI? I see more papers using VAR/VECM than XGBoost. Do you guys know why this?
0
votes
0
answers
41
views
Long-term trend prediction of time series data
I have a time series dataset project (single variable time series) on market share changes of a particular product in a region (values are recorded every day from 2018 to 2022) where I need to predict ...
0
votes
0
answers
9
views
Why is the conditional MSE equal to the MSE when the innovations of a time series are white noise?
I'm independently studying Lutkepohl's New Introduction to Multiple Time Series Analysis. In Chapter 2, we learn that the optimal (minimum mean squared error) h-step ahead predictor for a stable VAR ...
0
votes
0
answers
134
views
Statsmodels seasonal decompose interpretation
I'm currently working with a time series that has a monthly frequency. It has 3 years of data, 2019-2021. When using the statsmodels.seasonal_decompose I got the following plots:
I would like to know ...
0
votes
0
answers
13
views
Judge whether a time series of data is performing as expected or not
Given a dataset, where its variables change along the timeline. How can we tell the trend/performance is good or not?
Example:
id
timeline
F1
F2
1
[t_1, t_2, t_3, ...]
[a_1,a_2,a_3,...]
[a_4,a_5,a_6,...
1
vote
1
answer
17
views
Impact of residuals on forecast
I'm working with ARMA models right now and I was wondering about the following case:
If we have late significant lags in the residuals ACF and the rest of the earlier residual lags weren't significant,...
0
votes
0
answers
21
views
Time Series Forecasting with Different Time Horizon for Comparing Models?
At the moment I am dealing with a time series problem. The data I have is about 6 years and in daily frequency.
I want to try out different models on the data and I came up with an experiment:
...
0
votes
0
answers
67
views
How to increase the accuracy of a multivariate LSTM network?
I am using a pollution dataset with values for pm2.5, pm10 and pm1 as features and I am predicting the values for the pm2.5. I built an LSTM network but the predicted values are quite from the real ...
0
votes
0
answers
14
views
What social/economic/political domains are typ. easier-harder to forecast ? Are there benchmark estimates of accuracy for such time series forecasts?
For the report of the results from the Behavioral and Social Science Forecasting Collaborative Tournament I organized during the first COVID year, our team is trying to identify good benchmarks for ...
0
votes
0
answers
45
views
Estimate AR(1) parameters when DGP is AR(2) and the aim is a recursive forecast
I am exploring different options for making recursive forecasts for time series in the ML realm. I have found that it is easier to understand the basic concepts if I first use simpler models from the ...
1
vote
1
answer
39
views
Resources/books for project on forecasting models
My professor suggested a comparison of various forecasting models as a topic for my semseter project. Given that my only experience in statistics is the intro course in probabilty and statistics ...
0
votes
1
answer
32
views
Forecasting air passengers, but taking tickets sold into account
The most classic forecasting example seems to be something like: given historic data on the number of air passengers for each month up until say December 1959, predict the number of air passengers for ...
1
vote
1
answer
56
views
ARIMA accuracy measures, rolling forecast
Regarding ARIMA model selection and especially accuracy measures several questions came into my mind. To shortly summarize, in my understanding, after necessary transformations/differencing, p and q ...
0
votes
0
answers
68
views
Many-to-many time-series prediction problem
I would like some advice on how to implement an RNN or LSTM for my problem. I am working in Keras Tensorflow.
My data describes the moisture % histogram of a sample of material. There are 42 features ...
1
vote
1
answer
47
views
ARIMA: Understanding how time series analysis is focused on mathematical properties as opposed to best forecasts
Rob Hyndman states:
"The paper describing the competition [M] (Makridakis et al, 1982) had a profound effect on forecasting research. It caused researchers to: ... treat forecasting as a ...
0
votes
0
answers
17
views
Predictive model for a set (n=~250) of two related time series
I have a somewhat unique prediction/modelling scenario that I'm working on and need some guidance into appropriate modelling techniques. I have a set of approximately ~250 pairs of time series. These ...
0
votes
1
answer
70
views
SARIMA without seasonal differencing
I am trying to forecast daily data with (S)ARIMA, having observations for the last 180 days.
STL decomposition clearly shows seasonality and ACF plot shows spikes at 7, 14, 21, etc. days so that I ...
0
votes
0
answers
24
views
Looking for a intuitive explanation behind standard error formulas in linear regression
Suppose, we perform a linear regression for target variable Y with two features X1 and X2 (referred to as X below), I see that various standard errors are calculated as
$$ Mean\ residual\ error = MSE =...
0
votes
0
answers
41
views
Forecasting time series with shock
Goal: come up with the way to treat a shock in TS data to increase fit and quality of forecast
Given: Multiple time series monthly data for 6 years. There is a target TS with yearly seasonality and no ...
1
vote
1
answer
64
views
Journal reference for time series analysis and forecasting
What is the lay of the land of statistics journals that cover time series analysis? I would like to start reading journal articles on time series analysis and forecasting. My interest is in stochastic ...
5
votes
1
answer
335
views
How is time series analysis a different problem than forecasting?
Rob Hyndman states:
"The paper describing the competition [M] (Makridakis et al, 1982) had a profound effect on forecasting research. It caused researchers to: ... treat forecasting as a ...
0
votes
0
answers
14
views
Predicting 2 steps ahead with ARMA [duplicate]
How does one forecast 2 steps ahead in an ARMA model? If my training data is $y_1, ..., y_T$, and I have an ARMA(1, 1) model, then I have
$$\hat{y}_{T+1} = \alpha*y_T + \beta*(\hat{y}_T-y_T)$$
but ...
2
votes
0
answers
42
views
Time series: how much past predicts future
In financial (time series) statistics and forecasting we usually assume that the past of a series can predict the future to some extent. Every financial ad will warn you that investors should not ...
0
votes
0
answers
17
views
Weekly demand prediction of different product types at various depots
I have a dataset containing daily demand values of different product types at different depots. The products do not have the same demand and there are many cases where there has been no demand for a ...
1
vote
2
answers
120
views
How to deal with negative forecast values in time series forecasting?
Data - Monthly Rainfall of a region for the past 20 years
Objective - To Forecast for the next 2 years
I have used a SARIMA model and predictions have been done using R. but one of the forecast value ...
0
votes
1
answer
19
views
How to forecast a rental weekly sales demand using a 4 year history data?
I need to estimate the weekly demand required for a specific product in a specific week at a specific location.
I have the past 4 years daily data of each product at each location.
For example: number ...
0
votes
1
answer
32
views
Can time series forecasting be done without splitting the data into train/test sets?
The data is of monthly average rainfall for a specific region for the past 13 years (156 data points). What is the best way of splitting it into train/test sets?
I thought of selecting first 12 years ...
0
votes
1
answer
35
views
Make 21-days ahead forecast with daily log return data?
I want to make daily 1-days and 21-days ahead forecasts of a stock price. I have used daily log return data for both 1-day and 21-days forecasts.
Now I'm not sure if that is correct for the 21-day ...
0
votes
0
answers
74
views
How to forecast the trend of a time series based on external regressors?
Consider the following time series of sales in a store (black line) whose trend is highly correlated to the population size in the store nearby (red line). So, as the population grows also the sales ...
0
votes
0
answers
13
views
Choosing a validation and test sample selection for a re-forecasting model
I'm currently testing improvements to an income re-forecasting model and am unsure on how to select a good validation and test sample to consider the time element of the model. The model takes ...
1
vote
1
answer
67
views
Mysterious: linear model perfectly predicts sine waves forever ONLY when not in phase! [closed]
Overview
I am working with a very simple linear prediction model, and I am testing its limits using a basic example featuring two sine waves. I am shocked at how well the model is performing, and I ...
0
votes
0
answers
32
views
How to Model Constrained Outcome in Multivariate Regression Context
I have a problem in which the ratio of two different outcomes as a function of time $\frac{Y_{1t}}{Y_{2t}}$ is unknown constant $c$. I would like to estimate regression models of the following type:
$$...
0
votes
0
answers
73
views
Forecasting via residual bootstrap vs naïve in R
I'm trying to figure out how to apply residual bootstrap forecasting in R using an AR model and compare it to a naive approach that comes with the standard predict function in R.
BACKGROUND
Here's a ...
0
votes
0
answers
14
views
Can we do better than looking at an 7-day average of new Coronavirus infections? [duplicate]
If we look at the 7 or 14 day average of new infections, we are always lagging behind. I was wondering: can we do better? In essence, the issue with a normal 7-day average is that it gives equal ...
1
vote
0
answers
38
views
Calculations to forecast repayments on loans
I'm working on a dataset which includes the salaries of recent graduates in various job sectors. The data consists of students that have graduated in 2020 throughout the UK. I have calculated tuition ...
0
votes
0
answers
10
views
Creating a module to compare available forecasting techniques
I want to prepare a module for comparison of all forecasting techniques - beginning with classical approaches (arima, ses, also prophet etc.), then moving to sophisticated ML techniques (like LSTM, ES-...
0
votes
0
answers
17
views
Forecasting of ratios that need to addup to 100% [duplicate]
I intend to forecast 'shares' or ratios independently that need to add up to 100% . How can I do this especially when there are >2 ratios to predict. e.g. share of mdot, share of desktop and sahre ...
0
votes
0
answers
24
views
Confidence intervals for next frame video prediction
I'm using a ConvLSTM network for next frame video prediction. The output is a deterministic prediction of an image in the future. My question is: can a ConvLSTM model give me an interval of prediction ...
10
votes
1
answer
647
views
How well do Covid-19 forecasts work?
I hope this question is in the scope of CrossValidated (I think so because it is in the end about statistical analyses and machine learning, I am not looking for individual opinions but for reviews or ...
0
votes
1
answer
49
views
Forecasting future trends on electric vehicles
I'm new to forecasting and time series though I'm working on a dataset with the aim to calculate future trends of electric vehicles ownership. Based on this data:
...