All Questions
16,351 questions
1
vote
2
answers
1k
views
Lagged dependent variable with non stationary time series
I'm doing a regression analysis with non stationary time series. If I run the regression the residuals are auto correlated and non stationary. If i add a lag of the dependent variable (the estimated ...
CommunityBot
- 1
1
vote
1
answer
1k
views
Comparing trend/similarity among different time related variables
I have a list of stores and the important time related variables which affect the dynamics of the store in some way or the other. The aim is to generate stores which are similar to a target store. The ...
CommunityBot
- 1
3
votes
1
answer
567
views
How to build an adequate SARIMA model?
I'd appreciate some help with my data. I have to find (S)ARIMA models to fit some exportations data, but for some of them, I can't find an easy model by just looking ACF and PACF plots. So I change ...
CommunityBot
- 1
1
vote
1
answer
348
views
Is there a distance metric between ordered vectors?
I am looking for a distance metric between vectors whose elements are ordered,
i.e so the vectors:
[1,5,0,0,0,0,1], [1,0,4,0,0,1,0]
will be considered closer than
[1,5,0,0,0,0,1], [1,0,0,1,5,0,0]
for ...
CommunityBot
- 1
0
votes
1
answer
371
views
Removing autocorrelation when dependent/independent variables are unknown
I have access to forecasts produced using the methodology outlined in the following document, essentially to produce a forecast a linear regression against noon effective temperature (0.57 * noon ...
CommunityBot
- 1
1
vote
1
answer
2k
views
Lag/lead variables
I have a simple question related to the variables in a time series analysis.
I am using a panel fixed effect regression to see the impact of instrument issuance on firm performance. My independent ...
CommunityBot
- 1
4
votes
1
answer
2k
views
Seasonal component in daily time series remains after stl?
I'm following the procedure in this post (Adjusting daily time series data for the seasonal component) on R 3.2.3 (linux).
The de-seasoning process in the above post works fine. But with my data, a ...
CommunityBot
- 1
1
vote
0
answers
27
views
How to test if the frequency of events (gradient) changes over time?
I have time-series data in Python that tracks specific actions over time, with . I’ve quantified these actions and would like to test if the frequency of these events per unit of time (so basically ...
- 11
1
vote
1
answer
51
views
What's the intuition behind re-parameterization in the Dickey-Fuller test?
In text books and lecture slides, people often explain that the normal t-test of, say, the AR(1) parameter in the Dickey-Fuller test does not follow the usual distribution. It also explain that after ...
0
votes
0
answers
23
views
Time Series Decomposition Justification
Problem's background.
There is the Wold's theorem saying that any covariance-stationary stochastic process can be written as a sum of the infinite-order moving average model and deterministic ...
- 278
1
vote
1
answer
1k
views
Is there a formula for an innovation outlier regressor in time series intervention analysis
For an innovative outlier, the equations I got from a paper by Tsay, are these:
$$y_t = f(t) +z(t)\\
f(t) = w_0 \frac{\theta(B)}{\phi(B)}\epsilon_t^d\\
\epsilon_t^d =
\begin{cases}
1 & t = d ...
CommunityBot
- 1
4
votes
1
answer
4k
views
Relation between AR(p) stationarity and causality
Let's take an AR(p) model $\phi(L)y_t=z_t$ where $\phi(L)=1-\phi_1-...-\phi_pL^p$ and L is the lag operator. I have just studied that if there are no roots of the polynomial on the unit circle,
$1/\...
CommunityBot
- 1
3
votes
1
answer
472
views
Unsupervised Learning on Multilevel/Multidimensional Data
I am working on a case-control study, where I for each individual have high dimensional data (like illustrated in the image).
I would like to do both PCA analysis and Clustering on this data, but it ...
CommunityBot
- 1
1
vote
0
answers
28
views
Recursive one-step forecasting in timeseries model
I am trying to implement a recursive one-step forecasting approach for a Random Forest model.
The idea is to get a 12-months forecast in an iterative way where each prediction becomes part of the ...
- 140
3
votes
1
answer
27
views
I have several variables which sum to zero. Is it sensible to substitute each variable as the independent variable to get a better estimate of each
In economics there is an equation by Kalecki which sums over government expenditure, taxation, exports, imports, savings, consumption, investment, and profit to zero. The US NBER facilitates that with ...
- 86.5k
5
votes
2
answers
1k
views
Finding patterns in various-length binary sequences
I have a large number of binary sequences of different lengths (time-series of observations of the occurrence (1) and non-occurence (0) of some thing) and I am wondering how I could find patterns that ...
CommunityBot
- 1
1
vote
0
answers
23
views
Measuring correlation and percentage returns for data that can go both positive and negative [closed]
I am trying to measure the correlation between 2 random variables $X$ and $Y$. $X$ is current position of a stock, where $X>0$ would be long (buying) and $X<0$ would be short (selling). $Y$ is ...
- 69.5k
2
votes
1
answer
315
views
Prewhitening with seasonal response and non-seasonal independent variable
I'm working to develop a forecasting model for a quarterly seasonal variable (quarterly estimated individual income tax payments) using several candidates for non-seasonal independent variables (...
CommunityBot
- 1
2
votes
2
answers
57
views
Multicollinearity and estimation of correlation between time series in fMRI data
I have a fMRI data consisting of a set of time series for activity of each region of brain. There is a concept called functional connectivity which shows how activity of each region depend on other ...
1
vote
2
answers
413
views
Change detection in trending degradation data
I'm working with degradation data and are trying do use change detection methods to detect repairs. Since I'm looking for repairs I'm only interested in positive changes. Between the repairs the data ...
CommunityBot
- 1
1
vote
1
answer
541
views
Multiple Independent Variables, Multiple Dependent Variables and Time Series Data
I am conducting a research on stock market. My Independent variables are Oil prices, Exchange rate, Interest rate, GDP & Inflation. And Dependent variables are Market return, and Sector wise ...
CommunityBot
- 1
1
vote
1
answer
670
views
Blanchard and Quah output gap
In the paper of Blanchard and Quah (1989), they estimate the impact of demand shock on delta(real GDP). And from that they could estimate the impact of demand shock on real GDP (which is output gap). ...
CommunityBot
- 1
11
votes
3
answers
2k
views
Need advice on change point (step) detection
I have a time series with lots of steps/jumps (data file here). A plot is given below. I would like to subtract an appropriate value for each of these square wave features to bring them back down to ...
CommunityBot
- 1
2
votes
1
answer
64
views
Converting Monthly economic Data into daily data for Econometric analysis
I am trying to build an econometric model (VAR, VECM or regression) to predict daily interest rates based on daily financial data as well as monthly economic data.
I am unsure how to transition the ...
- 2,875
4
votes
1
answer
45
views
How to Simulate a Multilevel Predictor Variable with Both L1 and L2 Variance Components?
I'm working on simulating multilevel data where I have a predictor variable measured at Level 1 (L1), which has both L1 and L2 variance components. For example, I want to simulate a socio-economic ...
- 5,890
0
votes
0
answers
12
views
mice multilevel imputation: does specifying cluster variable ("-2" in predictor Matrix) without multilevel methods lead to cluster robust imputation?
In short: Are mice's imputations cluster robust when I only specify the cluster variable with "-2" in the predictor matrix but do not use multilevel models during imputation?
For clustered ...
- 1
1
vote
1
answer
2k
views
Time series split (expanding window) vs k-fold cross validation in time series forecasting
I do time series forecasting (q-o-q GDP) using ml models.
For hyper-parameter tunning I use grid search with cross-validation. Cross-validation is specifically Time series split (using expanding ...
CommunityBot
- 1
1
vote
1
answer
616
views
How to use the initial values computed for the Holt-Winters model to update the model to time t=n?
I have followed the technique for determining the initial level, trend and seasonal components for the Holt-Winters model as detailed by Rob Hyndman on https://robjhyndman.com/hyndsight/hw-...
CommunityBot
- 1
0
votes
0
answers
7
views
Cross-correlation SD
Is there a known best practice or mathematically well defined way of computing a continuous SD for cross correlation? Or do you just computer the SD at each sample point? I'm curious because then ...
0
votes
0
answers
17
views
Deriving a multiple based on actuals and forecast values
For context, we are using the DeepAR model for demand planning forecasting. Currently the forecast often underrepresents actual demand. It was suggested that we use a higher quantile to overestimate ...
- 101
3
votes
1
answer
514
views
Test to prove change in data structure over time?
I have a data of engine oil with its 18 parameters. We measure them every 1000km. It is being measured 11 times. At each measurement, each parameter changes.
How I can prove that the actual structure ...
CommunityBot
- 1
1
vote
1
answer
653
views
Hierarchical time series forecasting optimal reconciliation using Fable in R
I am doing hierarchical time series forecasting using fable package in R. I am using the optimal reconciliation method to reconcile the forecast. Here is the ...
CommunityBot
- 1
0
votes
1
answer
226
views
Bayesian Forecast: Credible interval with predicted regressors
I want to do a forecast of let's say orders with a Bayesian linear regression, where orders do not only depend on time but also on another regressor, let's say accounts at time t.
$$orders_{t} = \...
CommunityBot
- 1
0
votes
0
answers
15
views
Cross Correlation with Variable and Unsynchronized Sampling
I have two discrete signals that were not sampled at a constant rate and instead have a timestamp associated with each observation. The number of samples are large for each signal, but there are no ...
1
vote
1
answer
50
views
Why does FPCA not use scaling as PCA?
Functional principal component analysis (FPCA), according to the original paper, does not use scaling before FPCA, as in PCA. Instead, it uses a covariance matrix to compute the eigen-components.
I ...
- 46k
1
vote
0
answers
36
views
Why is Prophet faster than dynamic harmonic regression?
Section 12.2 of Forecasting: Principles and Practice (3rd edition) discusses Facebook's Prophet model. The authors write:
Prophet has the advantage of being much faster to estimate than the [dynamic ...
- 69.5k
1
vote
1
answer
462
views
Describe year-over-year survey data statistically when there are some same respondents but also new respondents
Assume a survey asks a question in year one that 100 respondents answer. In year two, the survey asks the identical question and 120 respondents answer, but only 80 of them took part in both surveys (...
CommunityBot
- 1
2
votes
1
answer
885
views
Stationary and non-stationary variables in time series - how to difference?
I want to predict a multivariate daily time series, the target output is the volume of packages that is send and the covariates are day specific information as weather, the distance to holidays but as ...
CommunityBot
- 1
4
votes
1
answer
64
views
Two sample difference of mean test when the samples come from time series
I have two sets of observations, A and B groups. I can use the t-tests to test the hypothesis that the mean of group B is larger than group A. But the inherent assumption here is that the samples in ...
- 46k
1
vote
2
answers
605
views
How to Recursively Predict a Time Series Using Neural Networks
I am currently using neural networks to forecast an electrical demand time series.
I am trying to create a forecast for the following day given previous observations at half hourly intervals.
My ...
CommunityBot
- 1
0
votes
1
answer
323
views
Conversion of ARMA to ARIMA in R using coefficient
I wanted to know given generated data from an ARMA(p,q) model, can I create data from an ARIMA (p,1,q) model?I tried in the web but unable to find anything .
For the ARMA parameters, choose μ,φ1,φ2,θ1,...
CommunityBot
- 1
5
votes
1
answer
3k
views
Using Rolling Forecast Origin Resampling in R for Neural Network Time Series
I am new to time series prediction and forecasting with neural networks and am having trouble with cross validation.
I am fitting a multivariate time series. I have 236 monthly observations. I am ...
CommunityBot
- 1
2
votes
1
answer
78
views
General formula for mixed models
I'm trying to wrap my head around the general formula of mixed models and how it relates to the system of equations I'm used to.
The general formula read like this:
$$\mathbf{Y_{j}}=\mathbf{X_{j} \...
- 153
4
votes
1
answer
655
views
GLMM with time-series covariance and binary response variable?
I have a binary response variable that was measured at irregular time intervals for a number of individuals. I want to fit a GLMM that accounts for the time-series covariance within individuals.
I ...
CommunityBot
- 1
2
votes
1
answer
229
views
Why are my predictied values from a Bayesian AR(1) model lagging behind the data?
Summary: I have simulated some data on an AR(1) process in R and fit the model in Stan. When plotting the predictions, the predicted values tend to lag behind the true values. Why is this?
Detail
I am ...
CommunityBot
- 1
3
votes
1
answer
367
views
Normalization within or across observations?
I am currently working on a dataset where the feature for each observation is time-series.
For instance assume two observations: "person X and person Y", and the feature 'price paid for milk' as ...
CommunityBot
- 1
1
vote
1
answer
238
views
Difference in results between the forecast function in R, and manually calculating the predicted value based on an ARIMA model
I have a series on which I fitted an ARIMA(4,0,4) model in R, and got the following estimations:
...
CommunityBot
- 1
3
votes
1
answer
3k
views
What number of lags for multivariate Portmanteau, Breusch-Godfrey, and Ljung-Box tests?
There are 3 types of tests for the residual autocorrelations here (I have a relatively small sample(58 obs):
...
CommunityBot
- 1
2
votes
1
answer
404
views
95% prediction interval for an ARMA(2,2) model
What would the formula for a 95% prediction interval for an ARMA(2,2) model be?
The specific model I am using is: an ARIMA(2,0,2) with non-zero mean, with the following parameter estimates:
...
CommunityBot
- 1
4
votes
1
answer
113
views
Do any time series models actually assume strict stationarity?
I am taking a Time Series course, and we have covered the topic of Stationarity. The following definitions are given:
Strict Stationarity:
A stochastic process $\{X_t\}_{t \in T}$ is said to be ...
- 5,130