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
0answers
9 views

How do they determine the “feels like” temperatures for weather data? [closed]

I'm pulling air temperature and other weather stats from various APIs. There's this concept called "feels like temperature", separate from "actual temperature". Apparently, this is the temperature ...
0
votes
1answer
11 views

Improving ARIMA forecasts

I am trying to use ARIMA to forecast stock returns. The problem is that variance is time-varying but ARIMA assumes that it is constant. As I understand, GARCH is only used for forecasting volatility (...
1
vote
0answers
8 views

R Dataset Tree Ring - Analysis ideas

I am trying to fit an appropriate model to the "treering" data-set (available in R). No simple ARIMA model seems to fit particularly well: for example, as remarked on p13 of http://people.stat.sfu.ca/...
0
votes
0answers
5 views

MASE Statistic: Comparing Across Time Series

I've read that the Mean Absolute Scaled Error (MASE) can be used to compare forecasting accuracy across datasets with different scales. I'm unclear on what this means. The MASE is a ratio of the ...
0
votes
0answers
7 views

Evaluating baggedETS on a Test set

I'm using the forecast package to test a variety of models on monthly sales for 400 products sold by my company. I'm following the practice of fitting on a test ...
0
votes
2answers
25 views

Time Series Forecasting with (week)day data in R

I was hoping to ask this community for a little advice. I’m doing a bit of daily forecasting, and, because it has always served me so well, I’m using Prof. Hyndman's now famous 'forecast' package. At ...
1
vote
1answer
17 views

RMSE value of LSTM model vs ARIMA model

I have a question that is a little bit confusing,, I have developed three models and one of them is LSTM,,, for the LSTM model ( I scaled the data -1,1 ) and run the model,, and to make a fair ...
0
votes
0answers
37 views
0
votes
0answers
12 views

Model to predict coronavirus (covid19) spread [closed]

im new in data sience and machine learning but i have some mathematical and statistics backgroud. I really just want some information about models (like papers or raw models). So if you have any ...
0
votes
0answers
12 views

Forecasting based on relationship between time series

I have 2 timeseries datasets let's call them A and B. So my problem is I want to forecast future values of timeseries A based on timeseries B. Timeseries A and B have the same characteristics i.e most ...
0
votes
0answers
9 views

LSTM and constraining input/output size

I'm new to neural networks but I've encountered a problem (which might be caused by my little experience with this topic). I would like to make a prediction of one measurement A in the future based ...
1
vote
1answer
14 views

Safety stock calculations using Intermittent Bootstrap

Together with fellow students, I’m working on an assignment to calculate safety stock levels in case of intermittent demand. We’re already able to simulate the demand like in the paper of Willemain, ...
2
votes
1answer
39 views

Forecast combination using optimal weights

I am struggling with a case where I am supposed to calculate optimal forecast weights of two forecasts. We have fitted the models on a training set (time series) and want to calculate optimal weights ...
0
votes
1answer
35 views

undifferencing seasonal time series

I have a daily seasonal time series of 1461 observations, which I seasonally differenced and I am running multiple different models. My trained data is of 1023 obs and the test data is of 438 obs. ...
0
votes
0answers
20 views

Help selecting ARIMA model whilst using ACF and PACF graphs

I have 20 years of quarterly data regarding investments. I am attempting to forecast future investment levels using SPSS. I have used seasonal decomposition to remove seasonal effects from the data, ...
1
vote
1answer
13 views

Looking at occurrences through a (daily) timeframe

How much distortion in our view can we expect when we look at data through a timeframe? For example, instead of stating the exact time of some item (e.g. 2020-03-23 2h33) we group and represent all ...
0
votes
2answers
58 views
+50

XGBoost one-step ahead forecast

I have trained and cross-validated an xgboost classification algorithm in R using the following code: ...
0
votes
0answers
14 views

Forecast regular values of differenced series

I have read How to “undifference” a time series variable but since it's been many years since it's been answered, maybe there's another solution. I have non-stationary series and I need to provide ...
0
votes
0answers
22 views

Can I include forecasted values as explanatory variables when forecasting?

I am wondering about a problem I have encountered. If I were to use income as an explanatory variable in an ARIMA model to forecast consumption, I would need future values for income to include in ...
0
votes
0answers
5 views

How do I create stress scenario of macroeconomic factor to compute Point in Time PD using Vasicek?

I have GDP data from 2009 to 2018 and have forecasted for 5 years till 2023. Using these forecasted GDP I can compute PIT PD. Assuming GDP and PD have negative correlation, I also want to create ...
0
votes
0answers
8 views

Pareto/NBD Seasonality Factor Zitzlsperger Implementation

I am trying to account for seasonality in my Pareto/NBD model based on the following paper. What I can't wrap my head around is equation 2b). My understanding of the seasonality is that it should be ...
0
votes
1answer
14 views

Forecast ARIMA and out of sample evaluation

I have a train daily data and test daily data. I fit an ARIMA to my train data, forecast it 7 days and I want to get some performance measures of the forecasted values. I can get performance measures ...
0
votes
1answer
33 views

Data science question: Time series analysis model comparison

How to compare two different forecasting models lets say one is the classical statistics-based model and the other is a machine learning-based or both from the same school of thought. Also, let's say ...
0
votes
0answers
10 views

Different ARIMA models for forecasting sales of many products, or one ARIMA model for all of them

I have sales data for many products, my task is to do an ARIMA model to forecast sales for each product. I've done a tailored ARIMA model for each product as they exhibit different patterns. Then ...
2
votes
1answer
61 views

how to model this multivariate time series?

Say I have a dataset as follows. ...
1
vote
0answers
17 views

Distribution of forecast error of ARMA + GARCH model

I am modelling a time series process and want to explore ARMA + GARCH. Using ARMA alone, with normally distributed residuals, we can determine the distribution of the forecast error using the ...
1
vote
1answer
57 views

Forecasting a time series $(x_t,{\bf Y_t})$ where all we care about is forecasting $x_t$

Consider a multivariate time series $(x_t,{\bf Y}_t)$ $1\le t \le n$ taking values in $\mathbb{R}^{d+1}$, and suppose that we wish to forecast $x_t$ using its own path as well as the "exogenous" ...
0
votes
0answers
16 views

How to forecast correctly the seasonality in R?

Consider the following code: ...
0
votes
0answers
19 views

Multi-Dimensional Time Series Forecasting - using model predictions as future input

I'm building a CNN that uses 10 different time series (all updated at the same time steps) to predict the value of another, 11th time series. What makes this problem a bit different from the typical ...
1
vote
0answers
22 views

R Forecast Package: Why are my in-sample simulations consistently higher than the original time-series when using the simulate function?

What I want to do: Fit an arima model with an xreg component using auto.arima and then generate simulations in-sample where I pass in different xreg values. The issue: After fitting an arima model (...
0
votes
0answers
25 views

R - How to create a forecast object from a pre-defined forecast, in order to apply accuracy() for MASE result

I am working with a forecast which has been created by Prophet. I would like to apply accuracy() to return the MASE after identifying this as the best accuracy measure for multiple forecasts, and will ...
0
votes
1answer
8 views

Forecast equation in Holt’s linear trend method

I have a question about the forecast equation in Holt’s linear trend method used to analyze time-series. As the forecast equation is prediction = l + hb where <...
0
votes
0answers
83 views

Suitable model to be trained on Chinese COVID-19 data in order to forecast other countries data

In an article by WHO (World Health Organization) published on February 24, they present a very interesting plot (page 7) comparing symptom onset vs diagnosed cases for China. Since there is a lag time ...
0
votes
0answers
18 views

TimeSeries vs Dataset with timestamp feature

I am not able to provide the exact values of the dataset due to data privacy issues. The variables I am using in my dataset are: Date (2007 to 2019), [A, B, C, D (Categorical Variables that doesn't ...
0
votes
1answer
42 views

I am trying to build a monthly revenue prediction model using Prophet in Python but how can I optimize the hyperparameters of the model?

The hyperparameters which I am trying to optimize are: 'n_changepoints', 'changepoint_range', 'holidays_prior_scale', 'seasonality_prior_scale' and 'changepoint_prior_scale'. 'changepoint_range' has a ...
1
vote
2answers
43 views

Forecasting 75 steps into the future ARIMA(1,1,1) Model EViews [closed]

Good afternoon, I have an economics time series of around 3300 daily observations that go over 10 years. I already developed and ARIMA(1,1,1) Model based on the autocorrelation and partial ...
1
vote
1answer
35 views

Time series forecasting when there is a capacity limit. Or in other words bounded forecasting

Ideas for dealing with bounding time series forecasting? My time series which are sales were limited for certain days in the past. For example, there was a capacity constraint on a day e.g. 1000 ...
0
votes
0answers
48 views

What's the relation between time series multi-step forecast with re-estimation and forecast errors obtained by tsCV()?

I have a monthly time series (yt) with 108 observations (1/2010 to 12/2018) which I divided into training [1:84] and test sets [85:108]. My idea is to train my model in the training set and apply it ...
8
votes
2answers
823 views

Choosing the right forecast model for exponential data (COVID19) forecast package R

I am trying to forecast aggregated daily COVID cases in Europe. These are present day numbers in Italy. ...
0
votes
1answer
66 views

When to use ARIMA model vs linear regression

I am trying to forecast time series of product sales, I started approaching the problem by implementing the ARIMA model, I iterated over all the possibilities of the models parameters (p, d, q) and ...
0
votes
1answer
78 views

time series for each customer to predict time to leave?

I am a beginner in the domain of forecasting and I was wondering if such a problem could be solved with time series analysis : given customer historical data of taxi pickups,along with the weather ...
0
votes
1answer
24 views

Determining order of ARIMA(p,d,q) from ACF and PACF

I know that when trying to determine if you have an AR(p) or MA(q) process, you look at the PACF and if it drops off significantly at a lag p, then you can say it's an AR(p), but if it's geometrically ...
0
votes
1answer
36 views

What is the take away from such ACF and PACF plot?

I plotted the ACF and PACF after taking the first difference of my time-series. What is the take away from such plots? What should be the AR and MA order in such case? The data is as follows, <...
1
vote
1answer
39 views

A solution to de-seasonalizing data? Admission rate to emergency room

I am trying to fit a model for the admission rate to our emergency department by using time series in STATA, which is fairly new to me and I would like to take your input about deseasonalizing. data:...
0
votes
2answers
52 views

sales data seasonality

This is the plot of my data and I'm wondering if these periods (on red circles) considered seasonality or not and if so how should I deal with them? can I model without smoothing them? Please ...
0
votes
0answers
11 views

Regression (linear/non-linear) to forecast daily sales trend

I want to estimate the daily sales on monthly basis, based on the last year's performance. My current methodology is: Generate monthly CDF (last day of the month will always be 100%); so I end up ...
1
vote
2answers
97 views

Working with Time Series data: splitting the dataset and putting the model into production

I've been working with ML for sometime now, especially Deep Learning, but I haven't work with Time Series before, and now I started working in a project for Demand Forecasting. I'm studying the ...
0
votes
0answers
15 views

What should I do when my ets forecast residuals are correlated?

I am currently forecasting seasonally un-adjusted GDP and ran into an issue with residuals where there is a very large negative residual around the financial crises of 07/08. As this is homework I ...
0
votes
0answers
14 views

What is a good train/test ratio for my time serias data?

I know this question is old and has no clear rule. But I will ask it for my particular situation and hope that I can get some general pointers. I have a time series with quarterly observations of ...
0
votes
0answers
16 views

Parametrization hourly data in cycle for in forecasting problem

I try to explain my programming problem. Dataset: 26 variables, 1176 hourly values Goal: Find optimal method to forecast next 24 values. I separated my historical data in training and test dataset in ...

1
2 3 4 5
58