Questions tagged [time-series]
Time series are data observed over time (either in continuous time or at discrete time periods).
1,872
questions
0
votes
1
answer
15
views
Out-of-Range Target Variable in Sequence-based Machine Learning Model
I'm encountering a scaling issue in a machine learning project. I'm predicting a target variable from an input sequence (and doing this for many). However, I've encountered a challenge where the ...
0
votes
0
answers
9
views
How to properly apply time series classification on a problematic data?
I would like to apply this work https://github.com/Vidhiwar/multimodule-ecg-classification/tree/master. In that GitHub link, (http://www.timeseriesclassification.com/description.php?Dataset=ECG5000) ...
0
votes
1
answer
43
views
Proper way to build time series dataset
I have daily time series but need to predict weekly total.
Here is example:
date
col0
col1
col2
col3
col4
target
0
2023-01-01 00:00:00
-1.021
0.708
-0.564
-0.502
0.244
-106.538
1
2023-01-02 00:00:...
0
votes
0
answers
25
views
Underestimation in the VAR model
I am trying to predict with a VAR model the number of orders of product X in company Y. It uses various indexes to do this. Unfortunately, the problem is that orders are specified in millions and such ...
1
vote
1
answer
45
views
How do I use ML models to estimate current stress level based on past data?
I am new to machine learning and I cannot understand the difference between estimating current stress level and predicting future stress levels based on historical data. I have been told these are two ...
0
votes
0
answers
12
views
Tensorflow RNN - implementing recursive layer
I am dealing with a regression problem, for which I wanted to try to use a recurrent neural network. The general setting is that I have to predict a continuous quantity starting from the evolution, in ...
0
votes
0
answers
22
views
Model for small sample time series
Im a total novice and i need to estimate some kind of relation-proof model(Granger test results, correl matrix are already provide some evidence) with following dataset:
20 observations (2001-2021), 4 ...
0
votes
0
answers
12
views
Is there any advantage to providing multi-dimensional input to torch modules?
Most layer types in torch.nn such as torch.nn.Linear accept input with more than one dimension. Is there any advantage in doing so if you can shape your data to represent a certain arrangement in ...
0
votes
0
answers
31
views
Winter Holt's Time series model
I am confused with the Winter Holt's Time series model usage. I use 2 years of data to train and want to predict 3rd-year data.
Note1: I have partial 3rd year's data, but I want to use it to check my ...
0
votes
0
answers
12
views
Clustering of two datasets in different years
I want to analyze two datasets by running a clustering algorithm on both and comparing the results. The two datasets have the same variables. The only difference is that one dataset is from 2010 and ...
0
votes
0
answers
10
views
Training multi-variate LSTM model with sample observations with differenet mean values
I am developing an LSTM model to predict the force-deformation response for wind turbine blades. I have generated the training data from a high-fidelity model for wind speeds ranging from 3m/s to 25m/...
0
votes
1
answer
44
views
Can lag features be applied into test data without label?
can lag features be applied into test data without label? I've been wondering. I tried to build random forest model using dataset: training data (with label Y) and testing data (without label Y). The ...
0
votes
0
answers
23
views
Prediction of Quarterly Financial Numbers
The task is to predict the quarterly revenue numbers using machine learning. Only 28 quarterly data points for financial numbers are available as companies release the revenue data quarterly. I have ...
0
votes
0
answers
48
views
Are there other pre-trained time series transformers than TimeGPT?
I am looking for pre-trained transformers trained on time-series data to use for transfer learning (for forecasting).
I found TimeGPT, and it does claim in this paper to be the first foundation model. ...
0
votes
1
answer
34
views
How can I approach this transactions data problem?
I am trying to approach the following problem: Imagine that I am a bank and I have a dataframe of transactions that customers make, the columns that this dataframe has are transaction date, customer ...
0
votes
0
answers
10
views
Stable test in online time series forecasting problem
I have a Time Series Forecasting problem. You can think of it as predicting the daily closing prices of Apple stocks. My data is divided into 4-day segments, and the forecasting is based on predicting ...
0
votes
0
answers
13
views
Autoregressive forecasting with distinct models problem
I got $n$ features - $f$ (used as an input as well as a target). Since I'm using linear regression and want to avoid situation in which weights of a model fit not only for $fi$ but for all of $f$ (...
0
votes
1
answer
24
views
Time Series Forecasting for Multiple Store Sales with Simultaneous Timestamps
I have a sales dataset with each store having a unique identifier. The dataset contains daily sales data for each store over a period of around two-years. I'm looking to build a time series ...
0
votes
0
answers
11
views
classification using simple relationships between time series data
I am looking to predict which courses are taught by which university professors at my school. More specifically, for each semester and professor I want to know the probability breakdown of which ...
0
votes
2
answers
60
views
Supervised or Unsupervised Learning Classification: Facebook Prophet vs. ARIMA
I'm currently exploring time series forecasting and considering the use of Facebook's Prophet and ARIMA models. I'm a bit confused about whether these approaches fall under supervised or unsupervised ...
0
votes
1
answer
24
views
Validity of using raw time series data for training of xgboost/random forest classifier
I am currently working on a project aiming of classification of process states based on time series data. For this, we are looking at different models, such as XGBoost-based classifiers or ...
0
votes
0
answers
7
views
Feature selection methods when input data is continuous but target variable is categorical
I plan on extracting features from a univariate time series, and use a feature selection method to select relevant features to predict a binary target variable via logistic regression. But, I have 2 ...
0
votes
0
answers
18
views
Predicting quanting sold using Time series data
I am struggling with a time series dataset comprising 12 features, including quantity sold and weather data, totaling approximately 1800 values. My goal has been to forecast future values, quantity ...
0
votes
0
answers
12
views
Validation loss hump in LSTM
I'm using PyTorch to fit an LSTM to a binary time series dataset which has about 300 time series of about 20 items. I am using 15% of the time series as a validation set. I then have an MLP on top of ...
0
votes
0
answers
9
views
Handling Irregular Time Series Data with Multiple Categories in Hourly Resolution Embeddings
I'm working with a time series dataset that has 100 categories for a single feature, and I've generated embeddings for these categories. However, my problem arises when I am trying to sample the data ...
0
votes
0
answers
19
views
LSTM - How can I predict the status an hour before in advance?
I’m very beginner, I’m trying to design a prediction model for forecasting the status one hour ahead.I have 150 sample data, each consisting of of 24 hours of time-series data with multiple features (...
2
votes
1
answer
113
views
Role of stateful parameter vs shuffle parameter in LSTM keras
I'm trying to make prediction on a multivariate time series using LSTM. I know stateful=True in keras LSTM means state(hidden) of each sequence, in a batch, at index i - is passed to the next batch, ...
0
votes
0
answers
34
views
Where to find the statement "time series discord is the best overall technique among all techniques" in Chandola et al.(2009)?
Senin et al. (2015) "Time series anomaly discovery with grammar-based compression" claim exactly the following (copied and pasted):
in a recent extensive empirical study by Chandola et al., ...
0
votes
0
answers
10
views
Kaplan Meier Estimator vs Weibull distribution for water pipe failures
I have a dataset of about 2.6k, featuring all the failures of the water pipe over 20 years. I have also added the right censored data, totaling the dataset (including failures) to be approximately ...
0
votes
0
answers
14
views
Models for multivariate time series with no feature that is known in future time points
I have and have only 3 columns in my dataset, which all are time series. I want to forecast one of them, and the other 2 are categorical, which are strongly correlated with the target and can be one-...
0
votes
0
answers
24
views
Forecasting using LSTM Model
I have a dataset with 12 variables (x1, x2, ..., x12) and one target variable (y). Is it possible to perform a forecasting for the target variable (y) over a certain period of time ahead without ...
0
votes
1
answer
36
views
Identification of inexactly-recurring material in time series stream
I am working on a personal project involving the analysis of a stream of audio data and the identification of (non-verbatim) repeated subsequences.
My research on time series has so far lead me to ...
0
votes
2
answers
134
views
What is the state-of-the-art in prediction\classification missing labels in partially labeled data?
Overview
Let's say I have the following data:
...
0
votes
1
answer
57
views
Prediction intervals for future timestamps - out-of-sample
I've created a model for out-of-sample forecasting that uses multistep recursive strategy to reduce my problem to regression, the predictions are sufficient but I was wondering if there is any ...
0
votes
0
answers
16
views
Good candidate time series models for 70-100 monthly data points, also incorporate past and future exogenous covariates
Per title, we're trying to identify a good time series modeling technique for:
70-100 variables of monthly sales or volume data (2015 or 2018 to present)
Ability to forecast not only using trend and ...
0
votes
0
answers
34
views
Using Bayesian statistics in time series forecasting
I would like to forecast demand count time series of taxi fleets at different locations on the map at different points in time. I.e. multivariate demand Time series forecasting.
Given hierarchinal ...
0
votes
0
answers
42
views
How to detect rare events in Time series?
I am working on a time series dataset in which each time step can be classified under 4 classes:
~EOI : P(~EOI) = .85
...
0
votes
0
answers
14
views
How to simulate from bivariate time series using VAR in R?
Background
I've run into a problem while trying to simulate from an existing time series using a VAR. My aim is to simulate future "trajectories" which I'd like to integrate in an existing ...
0
votes
0
answers
15
views
State estimation on live data
I have a time-series dataset that is poisson-distributed and each day I get a new datapoint. If I input all the data into a HMM (hmmlearn in python) it does a very good job at estimating the hidden (...
0
votes
0
answers
12
views
How to train model with data received from different 3x accelerometers sensors?
I want to make a model based on accelerometer data to recognise different activities like running, walking etc. I have a small dataset collected from my target sensor. I found another dataset with ...
0
votes
0
answers
13
views
Pattern recognition in time data serie
I have a log file with phone call data, described by 2 attributes: call_time, call_from. I want to discover some patterns in the data, using some statistical method. I don't have any predefined ...
0
votes
0
answers
12
views
Timeseries timestep changes in deployment
Im working with spatiotemporal data and I'm wondering if I train my model on a timeseries with a certain timestep, do I need to have the same timestep when I deploy the model and make predictions?
...
0
votes
0
answers
33
views
heavy underfitting of keras LSTM regression
I moved the question from stackoverflow to here.
I used keras LSTM to do the standard regression project of ...
0
votes
0
answers
11
views
How to dectect sudden change in signal frequency? (analysis of EEG signals)
I am trying to analyze data from EEG electrodes to understand how brain activity changes in different coginitive states (for simplicity, assume that there are only 2 states: baseline (A) and chanelled ...
0
votes
0
answers
35
views
3D CNN accuracy is too low, how to improve it?
I have just started learning image processing and this is my first time working on video classification. I am trying to develop a model that recognizes hand gestures using the EgoGesture dataset(more ...
0
votes
0
answers
65
views
a way to automatically split a video into chapters?
Given a video with audio, we can use ASR to get a script of the sentences and timestamps. We are looking for a way to group the sentences into chapters. There are several companies that are doing it ...
0
votes
0
answers
21
views
Why use sliding window input features in deep learning?
I was reading through the DNABERT paper and found that their input features were k-mers. This is equivalent to using rolling/sliding window features in the other common family of sequential problem, ...
0
votes
0
answers
28
views
Memory / Segfault doing auto_arima
I am trying to do a day based SARIMA modelling using the pmdarima package like so:
...
0
votes
0
answers
14
views
Comparing multiple multivariate datasets
Take the two datasets below:
default rate
state
age
income
asofdate
10
Texas
55
100,000
202309
14
Texas
35
97,000
202309
18
Texas
55
95,000
202308
22
Texas
35
95,000
202308
8
New York
21
55,000
...
2
votes
0
answers
44
views
LSTM output capped at a maximum
I am using a LSTM built using to forecast a single-value (solar irradiance) by using weather data as my input.
When predicting my validation test, I get a weird results as it looks like all my ...