Questions tagged [forecasting]
Forecasting is the process predicting future values based on historic and current data, typically for time-series datasets.
241
questions
0
votes
0answers
7 views
Using Multiple Features with HMM
I'm trying to model HMM with a dataset having multiple features along a time series.
Most ML models would operate on multiple features while most books carry example of HMM with 1 observed value and ...
0
votes
0answers
15 views
Very low error during training of a RNN for forecasting but high test error
I use a Recurrent Neural Network for time series forecasting of electrical load data from a cooling device based on past values of the load time series and temperature values. I first normalize the ...
0
votes
1answer
25 views
ML algorithm for high dimensional time series forecasting
I'm trying to make a forecasting model for goods prices in an economy (trying to forecast inflation).
Dataset: has 300 goods prices % monthly variations for last 6 years. And also added $n$ ...
0
votes
0answers
17 views
How can I use a prediction model (e.g., ARMA model or LSTM) for multi-variate data?
I have had a dataset below:
sensor1 sensor2 sensor3 ...
2021-01-01 1.32 2.2 1.0
2021-01-02 4.3 2.0 0.8 ...
...
I know ...
0
votes
0answers
62 views
Var_Imp Algorithms in Pred/Class Problems: Can I use it in TS Problems?
OBJECTIVE OF THIS POST:
Solve a query about the possibility of use prediction/classification variable importance tools in a time series type dataframe.
Collect the largest number of variable ...
0
votes
0answers
10 views
Prediction Intervals on (Multi-Step) Judgement Forecasts
Are there any R packages available or general methodologies for calculating prediction intervals on judgment-based forecasts?
I've looked at Hyndman's text and the R forecast package - which will ...
0
votes
0answers
13 views
identifying time series with threshold breach potential
(moved from stackoverflow.com)
Hi all,
I'm trying to solve a following problem. I have a set of various devices feeding their readings into a system where they are stored as time series: timestamp, ...
0
votes
0answers
11 views
Forecast methodology for geographic variables that are somewhat related
I'm creating time series forecasts for different geographies and wanted an expert opinion on how I can take into account geographic relationship to improve my model. Is there an algorithm that's ...
0
votes
0answers
12 views
What could be reasons for higher MAPE?
I built two models on a dataset where data for independent variables (X) being the same and dependent variable (Y) changes for each model
eg : Y is price value for a particular region. Y values change ...
0
votes
0answers
24 views
Getting KeyError while executing forcast( ) in statsmodel's holtwinters function
I'm trying to get time series prediction using the following code.
...
1
vote
1answer
10 views
How to deal with mismatch of timesteps in target variable and features in forecasting problem?
Background Info:
I am working with some climate data where I want to predict crop yields with my dataset containing climate- and satellite-derived features.
This is a time series regression ...
1
vote
0answers
13 views
Sarimax forecast : How to properly deal with non working days
(asked first on stackoverflow but felt like it would be smarter to put it here)
I'm trying to build a Sarima model to predict day by day the expected value of several measures (separately), the point ...
1
vote
0answers
27 views
Panel data classification and regression?
I am very new to time series/panel/longituginal data. From my understanding Panel data = multi-object time series and I have some panel data in long format (objects have multiple rows corresponding to ...
1
vote
0answers
14 views
Choosing kernel size of cnn for time series data with multiple seasonalities
I try to solve a standard time series forecasting problem using convolutional neural networks. The data has multiple seasonalities and so I wonder if a kernel size should reflect this fact e.g. for a ...
0
votes
0answers
16 views
Optimise for the sum of regression predictions?
I'm building a machine learning model to forecast the number of students on a course at a University. I'm currently optimising for MAE for each sample (i.e. a ...
1
vote
0answers
12 views
CoxPH model with Frailty and L1 regularization
This question stems from an approach proposed by Dr. Silverman, "Predicting Horse Race winners through A
Regularized Conditional Logistic Regression with Frailty."
In this paper, he proposes ...
0
votes
0answers
8 views
Is it possible to pass multiple features at once to Croston method?
I want to implement the Croston method for intermittent demand. I have a data frame that has 10 features and all those have many zeros in them. I want to pass the entire data frame to the Croston ...
2
votes
0answers
30 views
Time series forecasting for vibration prediction on Industrial machine production?
I'm working on a machine learning project related to an industrial machine.
The goal of the project is to build a model that would be able to predict the vibration of the machine while it's in ...
0
votes
0answers
32 views
Predicting Y Values Properly in a Regression Task using Scaled Values (Random Forest & MLP)
I have a supervised learning regression task: I am trying to forecast demand for a product based on sales in past years.
Data description:
Samples (rows) - Demand for a certain product (at a certain ...
0
votes
0answers
25 views
Time series forecasting in Python with 2 categorical variables
What approach is the best for a time series forecasting where you want to include 2 categorical variables in python? Im not finding any useful information that can help guide me with this; mainly ...
0
votes
0answers
9 views
Forecasting mon missing time timeseries data
I have time series data for minutes interval. But due to some noise i have to remove some rows from data. Now, I have data with some missing time stamp.
What should i do for forecasting in this case?
0
votes
0answers
11 views
In Time Series forecast, should Scaling be done on both train and test features combined ( test is 1 new data point)?
Let say I have a Time series, I'm using sliding/expanding window method to split to train and test data: train would be all the data I have until day x and test is day x+1.
To avoid Data leakage I'm ...
1
vote
0answers
46 views
Machine Learning Model for Time Series Forecasting
I am using Random Forest, SVM, and XGBoost models to nowcast/forecast an economic time series variable. However, I would like to extend these models to optimize/customize them for time series ...
1
vote
0answers
32 views
One Year Ahead Forecasting with Unevenly Spaced Time Series
I have many products in my warehouses which can be "demanded" any day by my different clients. I want to forecast how many of each item will be demanded for the whole next year. Naturally, ...
0
votes
0answers
6 views
Time series forecasting for stable pattern with some sudden changes
In my case, the time series is around a constant value with very small fluctuations. But, sometimes the signal starts increasing or decreasing for some duration of time. For my application, such ...
0
votes
0answers
21 views
I am building a anomoly detection model using Pyculiarity (Twitter's anomaly detection). However, the expected value is 'NaN'. What am I missing?
When I run the following code, I get the expected value as NaN:
...
0
votes
0answers
8 views
Time series forecasting when one of the series is known
I have a problem where there are two time series $\{x_t\}_{t \geq 1}$ and $\{z_t\}_{t \geq 1}$. These two time series are correlated for fixed time instant but uncorrelated with each other across time....
0
votes
1answer
125 views
Predicting out-of-sample time points with LSTM
I'm working on a time series forecasting problem using LSTM. The data is univariate and non-stationary. I followed this tutorial.
The data is processed as the following: First, the difference between ...
0
votes
1answer
31 views
Forecasting with Neural network and understanding which underlying model is favoured
If I have a very large set of data (~ 1TB). How can I use Neural Network on this data to understand which underlying distribution (eg. let's say a Gaussian or a Poissonian with a certain mean, sd) is ...
0
votes
0answers
20 views
Multivariate time series forecasting with the extended Theta method
I am looking for an implementation (better if in Python) of the extended Theta method for multivariate time series forecasting, presented in the following paper:
D. Thomakos, K. Nikolopoulos, ...
1
vote
2answers
347 views
Is time series forecasting possible with a transformer?
For my bachelor project I've been tasked with making a transformer that can forecast time series data, specifically powergrid data. I need to take a univariate time series of length N, that can then ...
1
vote
0answers
9 views
0
votes
0answers
24 views
Neural net performance using rmse
I am trying to build a NN which can predict exchange values. I am quite new to R and NN and I don't quite understand how I could improve the performance metrics of the neural network. I have tried ...
1
vote
1answer
51 views
Predicting high frequency sparse time series data in python
I have a dataset of a couple of EV charging stations (10 min frequency) over 1 year. This data consists of lots of 0's, since there is no continuous flow of cars coming to charge but rather ...
1
vote
1answer
27 views
Predicting sparse time series data
I have a dataset of a couple of EV charging stations (10 min frequency) over 1 year. This data consists of lots of 0's, since there is no continuous flow of cars coming to charge but rather ...
0
votes
0answers
25 views
Embedding a categorical variable and concatenating with a numerical variable, in a many-to-one sequence problem with multiple features
I have a small data set where I track 4 variables across 4 time periods, 1 categorical and 1 numerical variable. Below is picture of data set that I am using:
cat1 - Categorical variable encoded
num1 ...
0
votes
0answers
22 views
Azure ML / AutoML: problem with univariate time series forecasting
I'm having troubles generating univariate time series forecasts with Azure Automated Machine Learning (I know...).
What I'm doing
So I have about 5 years worth of monthly observations in a dataframe ...
0
votes
0answers
11 views
How can I intuitively calculate the accuracy of my financial prediction model?
I've built a SARIMAX model based on my personal spendings record as a college thesis and have reached a point where I'm pretty content with how it turned out and am getting ready to start writing the ...
0
votes
0answers
4 views
Supply Chain forecast with Covid-Data
I am working for a large food retail company and we are using ML models to predict the demand of certain products for the weeks to come.
Of course, looking at the sales distribution, 2020 was a ...
2
votes
0answers
115 views
Elman RNN with keras
I have to perform multi-step multivariate forecasting of time series, using keras. I found an example where LSTM is used. I could modify that example replacing LSTM with SimpleRNN.
Now I would like to ...
0
votes
0answers
21 views
Demand Prediction (Retail Sales) Datasets for Benchmarking
I've been looking for 2 days and can't seem to find what would normally appear as trivial:
a Time series (preferabl daily or intra-day resolution)
Demand or Sale Labelling
Retail Dataset
Preferably ...
0
votes
1answer
232 views
Problems to understand how to create the input data for time series forecasting with a recurrent neural network in Keras
I just started to use recurrent neural networks (RNN) with Keras for time-series forecasting and I found this tutorial Forecasting with RNN. I have difficulties understanding how to build the training ...
0
votes
0answers
6 views
Using forecasting values of wind speed at different hours in the future to predict power output at different hours in the future for a wind turbine
I need to design a neural network model for a wind turbine which takes as input the forcasting values of wind speed as follows : windspeedPlus001hr , windspeedPlus002hr , windspeedPlus003hr , ...
1
vote
0answers
24 views
How can i find the accurcy for time series? [closed]
I have to work for the first time with time series and I have some question about this interesting field of machine learning.
What I have to do:
I should make a forecast for quantity for some special ...
0
votes
0answers
13 views
Multidimensional Output from Radar Imagery and Climate Data
I am trying to predict what my rainfall field will look like at a future timestep using:
Radar imagery of rainfall fields at previous timesteps: A set of 2D matrices where each element in each matrix ...
1
vote
0answers
41 views
Time series forecasting. How use future values
I have a time series dataset containing hourly data from a few year, like below. Let's assume that I want to make prediction for the next 3 hours (2021-01-01 19:00, 2021-01-01 20:00, 2021-01-01 21:00)....
0
votes
1answer
139 views
Help improving time series prediction with LSTM on PyTorch
So, I am trying to use a LSTM model to forecast temperature data on PyTorch. I am relatively new to both PyTorch and the use of recurrent networks so I took a model I found on the internet to start. ...
0
votes
0answers
13 views
One LSTM for two currencies or two LSTM one for each currency?
Suppose I am building an LSTM model for currency forecasting. Assume that I am working on two rates: USD vs GBP and USD vs EUR.
Should I build one LSTM model with input size of two features (GBP and ...
0
votes
0answers
11 views
Forecasting as a supervised regression task, with uneven length timeseries - how to split?
Consider a number of timeseries. Here we have 3, just to make it dead simple. Note that they're all different lengths.
The very first thing I do is splitting by the original sample dimension:
Then, ...
0
votes
0answers
19 views
How to determine my pacf and acf in ARIMA with daily time series data?
I have daily time series data for revenue sales ranging from 2018-04-01 to 2021-02-19. I want to implement an ARIMA model and the plot of the data looks like this:
Based on my augmented-dickey fuller ...
