Questions tagged [forecasting]
Prediction of the future events. It is a special case of [prediction], in the context of [time-series].
3,323
questions
0
votes
1answer
24 views
How moving average works to make predicts?
A moving average model is:
Where $\epsilon_t$ is white noise that is $cov(\epsilon_t,\epsilon_{t-h})=0$, $var(e_t)=\sigma^2$,$E(e_t)=0$. How is it related to predict future by averaging previous ...
1
vote
0answers
11 views
Residual of ARIMA model
I've been using ARIMA recently so I'm a little bit unsettled in the chart interpretation. Currently I have 2 pictures of Ljung box and Residual in ARIMA (2,1,2). Can you help me comment on these 2 ...
0
votes
0answers
14 views
Supply Chain Spike Order Statistics
I work in supply chain. I am trying to identify orders that could be classified as spikes. I need to do this across many thousands of different products, each with vary different demand/order patterns....
0
votes
0answers
22 views
ARIMA total confidence interval for multiple predictions
I am using ARIMA (from statsmodels in python) to predict this years revenue based on data from the last ten years until year to date. For example now I'm predicting ...
0
votes
0answers
9 views
Extrapolation of total sales by partial stores results
My company receives the sales numbers from our different stores at different points in time over the next month. We would like to know what the total sales from all stores are while we don't have the ...
1
vote
0answers
21 views
Result of a diagnostic test of a predictive model looking too good
I have created a predictive model that outputs a predictive density. I used 1000 rolling windows to estimate the model and predict one step ahead in each window. I collected the 1000 predictions and ...
0
votes
0answers
12 views
Forecasting Incoming Payments - Increasing Variance in a Forecast
I am forecasting daily incoming deposits for a bank for the next 3-6 months. This is going to be used to forecast the bank's liquidity going forward. I have used both Holt-Winters and a time series ...
0
votes
0answers
5 views
How can we back transform forecasted series in ARIMA model using XLSTAT?
I am trying to back transform the forecasted rainfall series. How we can do it using XLSTAT Time series Box-Cox Transformation function.
1
vote
2answers
49 views
Recursive time series forecasting model
In a recursive forecasting model, let's say you are trying to predict sales of Target for the next month and you will append that prediction to your input and predict the month after. Basically, your ...
0
votes
0answers
24 views
Improving prediction of sub-daily library occupancy percentage with fbprophet
Out of interest for time series forecasting, I have compiled a dataset of the occupancy percentage of my university library over the last couple of weeks in 15 minute intervals.
Using this dataset, I ...
0
votes
1answer
17 views
Motivation behind GARCH
Suppose I have built an ARIMA model for a real life process, where volatility is present.
Now if for modelling the ARCH effect I fit a GARCH model, how will it affect my ARIMA model in terms of ...
0
votes
0answers
38 views
Recursive time series forecasting [duplicate]
In a recursive forecasting model, let's say you are trying to predict sales for the next month and you will append that prediction to your input and predict the month after. Basically, your target is ...
0
votes
0answers
21 views
Forecasting demand with stock-out
I'm using machine learning (gradient boosting) to forecast demand for a weekly magazine.
Every week, a new issue is shipped, while all stores wraps and returns the issue from the previous week. The ...
0
votes
0answers
18 views
Why my LSTM model predictions is a straigth line?
I am newbie in neural networks and I am trying to build a LSTM model to predict future values. My problem is that the plot of predictions result returns a line in comparation with the testting data.
I ...
1
vote
0answers
21 views
Assess calibration of a density forecast by Kolmogorov-Smirnov test on PIT of realized values
According to Elliott & Timmermann "Economic Forecasting" (2016) p. 429-430,
Calibration requires that if a density forecast assigns a certain probability to an event,
then the event ...
0
votes
0answers
9 views
Why is my multi-step SARIMA forecasting returning such good results?
I've been working on a SARIMA model that could forecast for up to 12 months into the future.
I'm using two functions to perform my forecasting.
The first auto_sarima...
2
votes
1answer
29 views
Is there a theortical benefit from increasing the frequency of a time series, eg monthly instead annual data
A model needs to be fit to a short time series of 3 annual observations of sales (2018 to 2020), e.g. 12, 14 and 13. The next 2 years (annual values) need to be forecasted bsaed on this data.
Is there ...
0
votes
0answers
6 views
Model that form relation between high frequency input and output variables, but observations are available at mixed frequency
I want to explore some forecasting models that form relation with daily output data and daily input data , but learns through monthly output data and daily input data. That means, output observations ...
3
votes
1answer
25 views
Random-walk and unit root processes predictable?
I know that a random walk is an AR(1) with a unit root, but there are also higher order autoregressive processes with unit roots. Does the unit root in such a higher order autoregressive process also ...
3
votes
1answer
95 views
Python: auto_arima predicts constant value
As a newbie, I am trying to implement the forecast using the auto Arima model. After searching, I found this site illustrates the usage and the hyperparameters used in the model. However, when I tried ...
0
votes
0answers
6 views
How to evaluate forecast results on holdout dataset for sales forecast for 300 stores?
I have forecasts for sales at 20 minute level, two weeks(december first week - high sales period and January second week - low sales period) for over 300 stores. There are Three models providing these ...
1
vote
1answer
31 views
Simple exponential smoothing with FORECAST.ETS in Excel: what are the calculations behind it?
I am really stuck trying to figure out why my manual calculation of ETS forecast doesn't match what's automatically produced by the FORECAST.ETS function in ...
0
votes
0answers
10 views
which kind of regression to apply in this case? (time series)
I have a daily time series data (independent variable) but the dependent variable only changes weekly. I.e if we have $N$ observations for the independent variable we only $N/7$ observations for the ...
0
votes
0answers
20 views
'low limit of 30% confidence interval' is higher than 'forecasts' in arimax model
I ran the ARIMAX model for multi-step forecasts (steps=n).
The code was written in python and I used the ARIMA model from statsmodels library.
(https://www.statsmodels.org/stable/generated/statsmodels....
0
votes
0answers
15 views
How do you test multi-step time series forecasting? [duplicate]
Suppose you have n observations of a time series dataset. You split it up to
n-k (train data) and k (test data) observations. You train a model using
the train data and you can now make predictions. ...
1
vote
1answer
35 views
Which time periods to use when calculating forecast accuracy for 12 month's rolling forecast?
I understand that the error in forecasting is the difference between the actual value and the forecasted value for time $t$. My confusion lies in which $t$ to consider for evaluating the forecast ...
0
votes
0answers
8 views
Creating Candidate Models for Forecasting using ACF and PACF
(Using R) I'm trying to find candidate models for forecasting, and I'm a bit confused. I used the auto.arima function on R to find the best fit model, which resulted with (i)SARIMA(1,0,3)(0,0,2)[12].
...
0
votes
0answers
30 views
Forecast is not deterministic for statsmodels statsmodels.tsa.statespace.sarimax.SARIMAX [closed]
I'm forecasting a data time series, but each time I do it, I get a different forecast, sometimes with variations of more than 5%, which is not considered acceptable.
The model Im using is v0.12.2 ...
0
votes
0answers
19 views
Identifying $ARIMA(p,d,q)\times(P,D,Q)s$ process
I have a series of monthly data (top left) differenced (12) with lag = 1 (bottom left) with the following ACF (top right) and PACF (bottom right):
And I'm trying to fit an $ARIMA(p,d,q)\times(P,D,Q)s$...
0
votes
1answer
30 views
Statistical forecast vs moving average at sku level
I have recently been doing some analysis regarding forecasting of specific items. I want to know if the findings of my analysis are specific to my data or if it is something that is commonly seen.
I ...
0
votes
0answers
11 views
Best way to handle special events at the end of each month in a time series
I am currently working with grouped time series where I have data disaggregated on a daily basis.
I use Facebook Prophet and the results are in general okay with one small catch.
There is one aspect ...
0
votes
0answers
9 views
Finding the Best Linear Predictor
Really easy question here!
Given the following model:
$X_t = \epsilon_t + 2\epsilon_{t-1}$
I want to try calculate/find
$Pred(X_t | X_{t-1})$
Here's what i've done so far
Find $X_{t-1} = \epsilon_{t-1}...
0
votes
0answers
10 views
What forecasting algorithm should I use?
I have a series of 2D euclidean coordinates. Around ~10-15, these represent player move destinations in a video game, I'm trying to predict future 2D points. I need an algorithm that can catch ...
0
votes
0answers
7 views
Predicted value smaller than the smallest value in train data in Tree algorithm
I ran the LGBM regressor in Python.
I thought that the predicted value (y_hat) from Tree algorithm should be in the range of train data (Y).
But I got a smaller predicted value (y_hat) which is ...
0
votes
0answers
9 views
Calculating 1 and step ahead forecasts
Given the following AR(2) Process
$X_t = 0.01+0.2X_{t-2}+\epsilon_t$
if $x_{100} = -0.01$
and $x_{99} = 0.02$ are the values of the series observed at times t = 100 and t = 99
How would I compute $\...
0
votes
0answers
14 views
Is SARIMAX dependent on the scale of an exogenous variable?
I built a SARIMAX model that forecasts some sales. The model includes one exogenous variable (the number of people on holiday). I used this implementation: https://www.statsmodels.org/stable/generated/...
2
votes
1answer
46 views
How to forecast from GARCH-copula model?
I am reading to understand how to forecasting time-series data from the GARCH-copula model. I am looking forward to understanding the steps. From my understanding, we should follow the following steps:...
0
votes
1answer
16 views
What is the best time series model to find “time lag” between 2 sets of data?
I have two sets of data.
Set 1 is a count of sales & date
Set 2 is a count of event x occurring & date.
What is the best method to find out how a change in an element of Set 1 (date t, sales s)...
0
votes
0answers
18 views
methods for hierarchical forecasting time series with non-equal length
Intermitent/noisy time series on the bottom levels of a hierarchy can be often dealt with a top-down forecasting approach. (https://otexts.com/fpp3/hierarchical.html)
The majority of top-down ...
1
vote
0answers
15 views
Best model or method for my time series data and question?
I have 2 sets of data
Set 1 is composed of time, t and number of u units , (sales data)
Set 2 is composed of time, t, and number of s sessions, ( count# of event s occured)
Things I would like to know:...
0
votes
0answers
20 views
Should you bin data when creating moving average to generate a forecast?
Let's say I have insurance claims that are created on a daily basis on Monday through Friday but sometimes there will be days in which we don't get a claim, perhaps it'll be a week or two that we don'...
1
vote
0answers
20 views
The Nonlinear Asymmetric GARCH Model
I'm reading about the Nonlinear Asymmetric GARCH (NAGARCH) model. If NAGARCH(1, 1) is given by:
$${\displaystyle ~\sigma _{t}^{2}=~\omega +~\alpha (~\epsilon _{t-1}-~\theta ~\sigma _{t-1})^{2}+~\beta ~...
2
votes
0answers
27 views
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 ...
1
vote
0answers
100 views
XGBoost for Time-Series Forecasting - Issues with Stationarity Transformations
I'm trying to forecast daily Covid vaccinations in Germany, especially focussing on using tree-based ensemble methods. One issue with tree-based methods for forecasting is extrapolation, when the time-...
0
votes
0answers
13 views
ETS model displays autocorrelation
How do you deal with an ETS model that displays autocorrelation? I would like to get rid of it but differencing so far has not helped.
12
votes
2answers
580 views
Why Time series decomposition is performed
I am new to time series forecasting. In most of the forecasting blogs that I have read so far, the time series is decomposed first. As per my current understanding it is suppose to help us in figuring ...
2
votes
1answer
37 views
Getting flat prediction with ARIMA model in Python
I am using an arima model to forecast sales of a given product in python, using statsmodels.tsa.arima.model.ARIMA Sales are daily, with a history of 2019 until ...
0
votes
0answers
26 views
Multivariate time series models for stock market prediction that utilizes the macroeconomics data
I wonder what Machine Learning models/library will work best if we want to build a forecasting system of S&P 500 index that is not just based on the daily market data, but also the macroeconomics ...
0
votes
0answers
11 views
Conditional Forecast for Time Series Model and the variance of forecast error
My Question:
Y(t) = 0.5 + 0.5*Y(t-1) + u(t), where u(t) is a white noise and is independent of I(t-1). Assume that Y(1) = 1, Y(2) = -1, your variance of the forecast error for Y(4) would be ____
My ...
0
votes
0answers
18 views
Modeltime R package forecast
I'm using modeltime package and try to understand from the tutorial https://business-science.github.io/timetk/articles/TK03_Forecasting_Using_Time_Series_Signature.html what the testing strategy is ...
