Questions tagged [time-series]
Time series are data observed over time (either in continuous time or at discrete time periods).
14,438
questions
2
votes
1
answer
653
views
Derive MA (Moving average) representation of a first-difference-process
I have a non-stationary AR(1)-process. After taking the first difference, how can I derive the MA representation of the resulting „difference process“ Delta_xt?
As an example, consider
xt = 1.5xt-1 ...
CommunityBot
- 1
0
votes
0
answers
5
views
Infra Down time prediction using ML
I have to predict the Infra down time for tenants hosted in multiple pods. I use signals like Average Page time, Application/DB CPU times, UI and other errors from the infra at a max(5min grain) or ...
- 11
0
votes
0
answers
10
views
Bootstrapping ridge regression in financial time series
I want to use ridge regression for a model of stock prediction, and want to use bootstrap to estimate the stability and variability of the coefficients. However, I have the concern that as financial ...
- 1
1
vote
2
answers
4k
views
Johansen Cointegration test in Python (statsmodels)
I have three time series df['a'], df['b'] and df['c'] which I want to test for cointegration ...
CommunityBot
- 1
0
votes
0
answers
20
views
Issues with Estimating Parameters in Gaussian Time Series Data Using R [closed]
I am working on a project involving time series data with Gaussian noise, and I am encountering an issue with parameter estimation. Specifically, my estimated parameter $\psi$ differs significantly ...
- 91
1
vote
1
answer
43
views
Should classical/traditional ML techniques such as polynomial regression/decision trees/random forests SIGNIFICANTLY outperform RNN in timeseries?
I have a dataset of numerous years of buoy wave height measurements including features such as measured significant wave height, numerical model predictions, peak wave period, mean wave period, and ...
- 65.7k
0
votes
0
answers
11
views
Time Based Cross Validation on market model
I am trying to predict the market price. My question is if it makes sense to use Time Based CV instead of traditional CV. Should I consider time while splitting the data to train/test/validate my ...
- 1
1
vote
1
answer
278
views
Correlation between previous year growth and monthly climate variable
I want to calculate Pearsons correlation coeff. between growth variable (tree-ring measurements) and monthly climate variable (e.g. monthly average temperatures) for a period of 100 years.
In that ...
CommunityBot
- 1
2
votes
2
answers
77
views
How does autoregressive training help limit compounding errors at inference?
I'm having a little trouble justifying something in my head and was hoping someone could provide some intuition? I understand for LSTM models or models that maintain some state about a sequence that ...
- 441
4
votes
2
answers
1k
views
How can I reduce the propagation of errors in multi-step time series forecasting?
I have a multi-step forecasting task where I am predicting values $H$ hours in the future.
Supposing that the forecast issue is done at time t, I will produce predictions for the next $H$ hours: $\{\...
- 441
1
vote
1
answer
593
views
How to handle outliers/spikes in time series and machine learning models when the number of observations is lower than expected?
I'm not that familiar with time series models, but I wanted some help to understand if there is a technique to handle outliers in periods where there are small number of observations.
For instance, if ...
CommunityBot
- 1
0
votes
0
answers
13
views
Proper detrend and anomaly method for monthly data?
I'm trying to calculate JJA anomalies of SST(Sea Surface Temperature) data.
before calculating the anomalies, i tried to detrend the data.
However, I am confused whether I should detrend June, July, ...
- 81.7k
0
votes
1
answer
222
views
Unconditional variance of MA(1)-GARCH(1,1) process
Let $y_t = \Delta{p_t}$ denote a time series of asset log-returns, where $p_t$ are logarithmic prices; $y_t$ is generated by the conditionally heteroscedastic MA(1) process
$y_t = \epsilon_t + \theta \...
CommunityBot
- 1
0
votes
0
answers
50
views
+50
What is backfilling when generating an AR sequence?
I am reading this paper: Extracting cycles from Nonstationary Data
I wish to recreate the Monte Carlo simulation in this paper.
I have a query prior to doing this.
On page 8, there is footnote 9, ...
- 265
1
vote
2
answers
909
views
Testing whether the correlation between two time series is constant through time
The Problem:
Given two time series, I want to determine if the (Pearson) correlation between the two time series is constant throughout time.
Example:
For example, I have the following two time series:...
CommunityBot
- 1
4
votes
1
answer
266
views
Stability of Time Series Hierarchical Clustering
We have a dataset with six time points and three biological replicates each. Therefore, we have a vector of 18 measurements for each feature, and used hierarchical clustering with Euclidean distance ...
CommunityBot
- 1
0
votes
0
answers
17
views
Apart from Recurrent Neural Networks what are some alternatives to AR(n), ARIMA(p,q) models?
I want to write a master thesis in CS (Machine learning). The topic that was assigned to me was time series prediction. I want to compare different ML methods with statistical methods (AR,ARIMA). I am ...
- 1
0
votes
0
answers
14
views
Strict white noise and weak white noise that is strict stationary implication
Let A is a strict white noise.
Let B is a weak white noise that is strict stationary.
What is more general, i.e. does A=>B or B=>A?
I know that strict white noise should have zero mean ad finite ...
- 173
1
vote
0
answers
14
views
Predicing next customers purchase dates (and possibly amount)
I need some help. I have a dataset with simple list of customer, date of purchase, amount. I'd like to predict the next purchase date for each customer and possibly the amount.
Customer
Date of ...
- 5,482
0
votes
0
answers
10
views
Structural Vector Autoregression: proof of the $(𝐾^2−𝐾)/2$ restrictions and identifiability?
I'm currently using Structural vector autoregressive models by Kevin Kotzé to learn Vector Autoregression. One of the points that it makes is the following:
the number of restrictions that we need to ...
- 68.8k
1
vote
1
answer
1k
views
Log transformation of TS-stationary time series?
I have another question about main econometric time series transformation.
I usually see the $log$ transformation of prices: $$p_{new}\left(t\right) = \ln\left(\frac{p_t}{p_{t-1}}\right), t \in [2\...
CommunityBot
- 1
1
vote
0
answers
32
views
Overfitting Time Series
I have only one time series $(y_0, t_0), (y_1,t_1), \ldots, (y_n, t_n)$, with $y_i \in \mathbb{R}$ and $t_0 < \cdots < t_n$. The believe is that these are points on a function $f(t; \mu)$ with $\...
- 572
1
vote
1
answer
337
views
P-value correction for multiple Mann-Whitney tests, some of them being dependent
I have performed multiple comparisons using Mann-Whitney U tests, and want to correct the p-values to know which results are worth reporting. The structure of the data is as follows :
2 experiments, ...
CommunityBot
- 1
0
votes
0
answers
10
views
Looking for Composition Differences in a Multi-Year Dataset
I have a multi-year dataset in which I am interested in a genotype (which I can code as a factor) over time for a particular species. I am interested in if there are significant differences in ...
0
votes
0
answers
23
views
Transferring a regression with ARIMA error model from R to Excel (formula help)
I'm trying to translate an ARIMA model into an Excel spreadsheet for calculations for something I'm studying.
The model is a ARIMA(4,1,0) that I used the ...
- 68.8k
0
votes
0
answers
16
views
How to preview plots in R Notebook? [closed]
I'm trying to plot a correlogram and periodogram for my dataset in R Notebook, as if I understand correctly, R Notebook offers the feature to preview plots without having to knit unlike R Markdown. ...
- 43
1
vote
1
answer
850
views
Time series data with a independent variable
I'm working on a time series data in python, which has sales at a day level. As expected, there a few peaks at holidays and long weekends. I would like to add a new binary class variable "Holiday&...
- 1,005
0
votes
0
answers
5
views
Which analysis is suitable for comparing cross-correlation coefficients between experimental phases in a repeated measures design?
I have a repeated measures experimental design, with a within-subjects factor phase and a between-subjects factor order of phases. I will compare pairs of time series data (for pairs of participants) ...
- 9,597
3
votes
2
answers
247
views
Mean square convergence of a series of stationary random variables
In Brockwell and Davis's book (Time Series Theory and Methods 2nd Edition), there is the following problem:
Show that if $\{X_t, t=0, \pm1, \dots\}$ is weak stationary and $|\theta| < 1$ then for ...
- 1,005
9
votes
1
answer
2k
views
PyMC3 implementation of Bayesian MMM: poor posterior inference
Google released a whitepaper on Media Mix Modelling (MMM) in 2017; vanilla MMM (established in the 1960s) uses multivariate regression. It's a decent mechanism to understand which of your marketing ...
- 9,597
0
votes
0
answers
9
views
PCA with Time Series and Leakage
I have the following code for a simple statistical factor model with PCA and moving window for time series data:
...
- 1
0
votes
1
answer
15
views
Is an ARCH model regressed over previous time series values or their residuals?
Suppose we have a time series $\{y_t\}$ which we would like to model using an ARCH or GARCH model. That is we assume the time series can be written in the form
$$y_t = \mu_t + \epsilon_t$$
where $\...
- 68.8k
0
votes
0
answers
8
views
Difficulties in using exogenous variables for time series forecasting
I have a slight confusion when time series modeling using exogenous variables. Suppose I am modeling housing price, and one of my exogenous variables is 'housing type' with value 'unit' or 'house'. ...
- 11
0
votes
0
answers
30
views
Unit Root tests for Timeseries Analysis - Derivation of the Characteristic Equation [duplicate]
The below equation describes a timeseries process.
$${\displaystyle y_{t}=a_{1}y_{t-1}+a_{2}y_{t-2}+\cdots +a_{p}y_{t-p}+\varepsilon _{t}}$$
To determine whether this process contains a Unit Root, we ...
- 330k
0
votes
0
answers
11
views
Good method for sorting out samples with/without oscillation pattern in panel data?
I have biomed panel data. Some variables are time series (blood tests from ~320 participants) with few samples (~20-100), others are fixed over time (age, height...). There is an intervention at some ...
- 319
2
votes
1
answer
911
views
Maximum Likelihood Estimation of a VAR(p) Process
In Lütkepohl's New Introduction to Multiple Time Series Analysis Chapter 3, we try to estimate the model parameters via Maximum Likelihood Estimation.
We assume that $u=(u_1',\dots,u_T')'\sim \mathcal{...
CommunityBot
- 1
1
vote
1
answer
24
views
Estimating a cointegration vector using OLS
I have been reading Hamilton, and just came across the following concept.
To estimate the cointegrating vector, first set the first component to 1. Then, perform a linear regression to estimate the ...
- 2,667
0
votes
1
answer
567
views
Ljung Box test for residuals of constrained ARIMAX(2,1,0) model
I have this ARIMA(2,1,0) model with one exogenous variable: $$\Delta y_t=c+\phi_2 \Delta y_{t-2}+\beta_x x_t+\varepsilon_t$$
I want to run Ljung Box test of residual autocorrelation with test ...
CommunityBot
- 1
3
votes
0
answers
193
views
Several python libraries for causal impact analysis based on R causal impact but which one should I use? [closed]
My question: I liked the Google causalimpact package in R and want to do the same work in Python. Which package should I use?
The very first library I saw is a port of Google's R causal impact package,...
- 179
23
votes
3
answers
42k
views
Which Dickey-Fuller test for a time series modelled with an intercept/drift and a linear trend?
Short version:
I have a time series of climate data that I'm testing for stationarity. Based on previous research, I expect the model underlying (or "generating", so to speak) the data to ...
- 81.7k
4
votes
1
answer
1k
views
Loss function for vectors when magnitude of elements and their position are both important
Context:
I am using a transformer for time series prediction.
The target and predicted tensors are both of size (8, 10, 181) which represents (batch_size, number of ...
- 1,005
1
vote
1
answer
371
views
Recommendation for fitting mainly linear time-series data to a curve
Most of the time-series data I'll be looking at is linear and uniform - a straightish light parallel to the x-axis.
The exceptions I need to find are those that deviate recently, the last one, or two ...
CommunityBot
- 1
1
vote
1
answer
13
views
Title: Issues with Fitting a Quasi-Periodic Function with a Trend Term in Python
I'm trying to fit a quasi-periodic function with a trend term to my time series data using Python. The data consists of monthly observations from 1963 to 1976. My goal is to model the data with a ...
- 805
0
votes
0
answers
9
views
Model binary-choice decision from time-series data
I am looking for some suggestions for modeling the following decision-making problem:
Subjects had two choices when making a decision in a randomized control trial. The decision-making process can ...
- 67
2
votes
1
answer
340
views
Discovering peaks/patterns in time-series and clustering them
I have a dataset which contains minute level sensor measurements. Sample is shown here:
To me useful information are these peaks in time series, mostly their peak and duration. My idea is to take out ...
- 1,005
0
votes
1
answer
49
views
Estimation of model coefficients of ARIMA model
Let say I have below ARIMA model estimation in R
...
- 508
0
votes
0
answers
20
views
Alternative to random intercept cross-lagged panel model
I'm looking for some inspiration as how to best model a given phenomenon, thought I'd ask around here.
This is a study where we will be collecting 5 repeated measures (say blood pressure, BP) over the ...
- 68.8k
0
votes
0
answers
7
views
Panel time series methods and endogeneity
Do panel time series methods such as Pooled Mean Group Estimator rely on an exogeneity assumptions of the explanatory variables or does the presence of a cointegrating relationship guarantee that ...
- 68.8k
4
votes
1
answer
89
views
How do I test if a population change is statistically significant?
I have data showing different statistics for all parishes in the country for 2014 and 2024 (I actually have data for all the years in between as well, but I am comparing these two years). The data ...
- 19.8k
0
votes
1
answer
273
views
Are the forecasting methods like mean, naive, drift, weighted average applicable to non stationary time series?
Like AR, MA models essentially need the series to be stationary, do the other forecast methods mentioned above also follow stationary?
CommunityBot
- 1