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
0
votes
2answers
25 views

Which Nonparametric Model to use for Small Time Series?

I have the following data: ...
0
votes
0answers
25 views

Long-term forecast based on ARIMA with external regressor: Correct approach? Violation of assumptions?

I'm new to time series analysis and involved in a project of forecasting sales of a certain product. Here's a quick summary: There are around 15 years of historical sales data available, which are ...
1
vote
0answers
15 views

Should I always convert time-series data to stationary before forcasting?

I am trying to predict how much revenue a store will generate in next month based on revenues of previous months. I was doing simple regression for forecasting before, but I have recently read about ...
0
votes
0answers
11 views

Modeling multiple time series variables for per/day observations under 2 periods

I'm trying to use a twitter volume per day variable and a google trends frequency per day variable to predict price of a volatile crypto currency. More specifically, I'm trying to replicate the model ...
0
votes
0answers
15 views

What are issues or best practices related to using fitted values from one model as predictors in a second regression forecast model?

I am creating two models, where the fitted values of one SARIMAX type model are used as an exogenous predictor for a second dynamic regression model. I want to understand what assumptions of the gauss ...
0
votes
1answer
17 views

Forecasting count data after fitting a Poisson regression model

I have created a Poisson regression model in Python for predicting the number of orders a company will receive a day based on a dataset of order counts and a number (~5) factors that contribute to ...
1
vote
0answers
23 views

Univariate, Multivariate, Cross-Sectional, Repeated/Pooled Cross-Sectional, Panel or Longitudinal Analysis?

I am wondering if there are formal definition that will help distinguish between univariate, multivariate, cross-sectional, repeated/pooled cross-sectional, panel and longitudinal analysis? ...
1
vote
2answers
28 views

Can stationarity occur over different time periods?

According to what I've read on stationarity it seems like it is either an all or nothing property. Basically, if we have a time series with a unit root then it progresses as a random walk with no ...
0
votes
0answers
68 views

What is the best method to predict the water consumption of EACH customer for the next month?

Say we have a dataset that has the following attributes: customer_id: There are a total of 1000 customers, each of them with a unique customer_id observation_date: The date on which we last observed ...
1
vote
2answers
32 views

Estimate Unique Number of Visitors

Is there a way to estimate the number of unique monthly visitors to a site based on a limited sample of one week of data? I have information about when a given user visited the site. This isn't as ...
0
votes
0answers
9 views

Equivalence between ARIMA and HMM

The question is about the equivalence between ARIMA models and hidden Markov models in the context of time series analysis/prediction. Specifically: Can any ARIMA(p,d,q) model bet represented by an ...
0
votes
0answers
16 views

Have I seemingly constructed a valid COVID-19 nearest-neighbor state forecast? [closed]

There is literature supporting nearest-neighbor model forecasting based on several favorable attributes. The first of which is apparent flexibility in choosing what “near” means in nearest neighbor ...
2
votes
0answers
17 views

What are the underlying statistical differences between the vector autoregression model and Prophet?

I am trying to understand the underlying fundamental/statistical differences between vector autoregression models and Facebook's Prophet, with regards to multivariate time series forecasting. I am ...
1
vote
1answer
25 views

Regression on small values

I am working on a problem to predict a stock's returns with a certain set of features. The problem which I am facing is that when I try to predict the stock price itself, the models capture the trend ...
0
votes
0answers
37 views

Forecasting hierarchical time series

I have to forecast overall monthly sales in different shops in a bunch of cities. The structure of my dataset is hierarchical - I have a list of products and their monthly sales for every shop. The ...
1
vote
0answers
19 views

Difference between forecasting and predicting in statsmodels SARIMAX

I am using SARIMAX model from the statsmodels library to predict(forecast) future values in a time-series. The library contains four methods: predict(), get_predictions(), forecast(), get forecast(). ...
1
vote
1answer
15 views

forecast of a time series model by taking into account over/under cost

Let $T$ denote a univariate time series data, and $c_1$ represent the cost of one unit of over forecasting and $c_2$ represent the cost of one unit of under forecasting. Suppose we can estimate $c_1$ ...
0
votes
0answers
14 views

Possible to make predictions with overlapning data

I've the following dataset. The task is whether "UNEMP_Q1" is correlated with "UNEMP_AVG_Q47" and whether it is possible to use "UNEMP_Q1" (or/and UNEMP_Q2 and UNEMP_Q3 ...
0
votes
0answers
26 views

What model for highly irregular time series and very few observations?

I would like to apply an ARIMA model to some data to forecast values into the future. It has to do with bank disbursements. The problem is that, for a normal project, you may have up to 8 observations ...
2
votes
0answers
12 views

Application of xreg in forecasts

I'm trying to compare my forecast values with real values and assess the difference. I have 130 observations which are number of deaths (dependent variable). As well I have data for interaction ...
0
votes
0answers
26 views

How to calculate the MSE of exponential smoothing model in R?

I have a dataset of the daily stock price and volume, like this: ...
4
votes
1answer
52 views

How can proper scoring rules optimize the probabilistic prediction compared to improper scoring rules?

I understand the fundamentals in the decision theory about accuracy being an improper scoring rule compared to other proper scoring rules like ...
1
vote
1answer
19 views

Prediction model of test scores based on subjective assessment

I am trying to build a model where I want to measure the accuracy with which supervisors can predict the outcome of test taker's scores. For example, supervisors rate test taker's subjectively before ...
6
votes
3answers
173 views

Time series forecasting: from ARIMA to LSTM

I am looking for resources on the techniques for time series forecasting. It seems that there are three approaches, listed below in the order of their machine learning-ness (and correspondingly their ...
0
votes
1answer
25 views

How to forecast with certain conditions

I have a doubt about forecasting analysis. Let's say I have a "time" variable and a "budget" variable. I would like to build two model: time series model where I forecast the ...
0
votes
0answers
19 views

How to predict unknown time series in using Facebook Prophet?

The problem is predicting hits on different stories published on a website. I am aware that for this kind of time series forecasting, facebook prophet is a popular library. However, it seems in ...
0
votes
0answers
22 views

forecasting monthly outcome with quarterly data - ARIMAX in r

I am new to R. I got quarterly data of one time series x1 and monthly data of the exogenous variable x2. I try to forecast the next three month of x1 using an ARIMAX model in R. I came up with this ...
0
votes
0answers
11 views

Modeling different days using Prophet

My time series dataset has varying data depending upon the day of the week. During Fridays, Saturdays and Sundays, the values are abnormally low. However, from Monday to Thursday, they maintain their ...
3
votes
2answers
50 views

Unable to recreate Statsmodels ARIMAX (1, 1, 0) forecasts by hand

I have fit an ARIMAX (1, 1, 0) model to a timeseries dataset consisting of 1 endogenous timeseries ("Y") and 1 exogenous timeseries ("X"). My exogenous timeseries in the model was ...
0
votes
0answers
9 views

Insufficient n > p for forecasting with an RNN

What is the best approach to handle insufficient sample size when forecasting for multiple sequences simultaneously with an RNN? My training set has n=956 (time points) and p=262 (sequences). I'm ...
3
votes
0answers
28 views

Multivariate time series forecasting with trend and seasonality

I'm working on a dataset which has city and product level sales for the past 3 years - on a daily level. My objective is to forecast the sales for the next 3 months - for a given city, product and ...
1
vote
0answers
20 views

Forecasting with extrenal regressors in R using RUGARCH

I am struggling to find the solution of my problem, I want to model the volatility of the DAX index using some explanatory variables to do so. I am using the rugarch packed and I model the series has ...
1
vote
0answers
30 views

Forecasting with VAR model

Suppose we want to build a VAR model with two non stationary historical series $\{X_t\}$ and $\{Y_t\}$ . Let us further suppose that in orded to get stationarity I should tranform the series ...
1
vote
0answers
11 views

State Space model code from Time Series analysis Springer book, and code error from optimization

I'm trying to apply State Space model code from the book, Time Series analysis and its applications from Springer book, to my data. The code is as below: ...
1
vote
1answer
34 views

Why do time series, with positive values decomposed into seasonal plots, have negative values

I have been running different forecasting algorithms such as Facebook Prophet and Forecast (from R) and I note that despite all my time series values being positive, my seasonality values are negative....
0
votes
0answers
11 views

How to represent dependencies between horizons and uncertainty when forecasting multiple horizons?

Suppose the following case of univariate time series forecasting. We forecast sales of a product 1 till 7 days ahead. The true distribution of the actual sales is the following: there is a 100% chance ...
0
votes
1answer
37 views

Forecast confidence intervals from multiple realizations

I have a forecast which involves sampling a probability distribution and therefore each time I run the forecast there is some random variation between results. If I run the forecast many times, how do ...
0
votes
0answers
22 views

When need to detrend the time-series data?

I am new to time-series models. I read many books, tutorial, relative questions, and published papers. However, I still confused about the detrended term. I would like to know when I need to detrend ...
-1
votes
0answers
17 views

How would you convert an ARIMA(0,0,1)(0,1,0)12 model to equation form? [duplicate]

How would you convert an ARIMA(0,1,1)(0,1,1)12 model to equation form?
1
vote
0answers
20 views

Forecast based on indicators

I wanna forecast a specific timeseries. I have monthly datapoints for about 8 years. (From now on called target). In addition to that i have 15 indicator timeseries (also monthly with about 8 years of ...
0
votes
0answers
7 views

Accuracy of sliding window in ARIMA models

I have 245 daily stock returns with some explanatory variables. I will compare the ARIMA and ARIMAX models with some basic machine learning techniques. The training periods will be 30, 60, 90, 120 and ...
1
vote
1answer
37 views

Python: pmdarima, autoarima does not work with large data

I have a Dataframe with around 80.000 observations taken every 15 min. The seasonal parameter m is assumed with 96, because every 24h the pattern repeats. When I insert these informations in my ...
0
votes
1answer
18 views

Forecasting what market share will be for a basket of products after a strategy has been implemented?

I work for an online company which has the ability to shift market share for group of product in a market baskets i.e. frying pans. In the market basket for frying pans there are 5 products and of ...
0
votes
1answer
13 views

Rolling Time Period for Exponential Forecast (Holt Method)?

I am curious in which cases (if any) you would force an exponential smoothing function to only incorporate data from the past year? Per Holt method, one would continue to use the full time series as ...
0
votes
0answers
19 views

Choosing between additive and multiplicative Holt-Winters Model

While using the Holt-Winters model for seasonality, I am unable to choose a better fit between additive and multiplicative models. I used to look at RMSE value and choose the one with the lower RMSE. ...
0
votes
0answers
10 views

Time Series Normalization by row or by columns?

In times series forecasting when I perform a "global training" learning from many time series simultaneously for example (SVM regression, Neural Networks and etc..). I can normalise each ...
2
votes
1answer
18 views

How to fit an ARMA model along with “usual” predictors?

Is there an R package that I can use to fit a model being a blend of the ARMA model and the ordinary linear regression? I mean something like this: $$Y_t=\beta_0+\beta_1 X_1+\beta_2 X_2+\beta_3Y_{t-1}+...
1
vote
1answer
39 views

Extract the distribution parameters from a forecast that produces a fable

I have used the forecast(h="1 year") function from the fabletools package to produce a forecast from an ARIMA(1,1,3)(1,1,0)[52] model where the dataset had been transformed. A snapshot of ...
3
votes
3answers
147 views

Permutation feature importance on Train vs Validation set

When I compare on Permutation Feature Importance (PFI) on Train vs Validation set, some features has high values (of PFI) for train but the low values (PFI) for validation. One the conclusion, for me, ...
1
vote
0answers
25 views

Pros and cons of converting weekly to daily data

I am trying to forecast an economic variable called the "yield spread" in python. Among the variables in my dataset, two of them are measured on a weekly basis. These are: unemployment ...

1
2 3 4 5
62