Skip to main content
Cross Validated

Questions tagged [autocorrelation]

Ask Question

Autocorrelation (serial correlation) is the correlation of a series of data with itself at some lag. This is an important topic in time series analysis.

1,776 questions
Newest Active Bountied Unanswered
Filter by
Sorted by
Tagged with
1 vote
0 answers
25 views

I am working on my undergraduate thesis. I'm working with a balanced macro-panel of N = 17 (countries of the same region) and T = 32 (years). Considering that my panel shows a certain level of ...
David Quintero's user avatar
  • 11
0 votes
0 answers
44 views

I am running a GAMM model to look at the effect of temperature on the daily patterns (in radians) of an animal's movement (step length). The data is hourly, however there are individuals with hours, ...
Amelia's user avatar
  • 31
3 votes
1 answer
41 views

I have relational data, i.e. observations for pairs of objects. More specifically these are migration rates between plant populations, which I would like to explain by a predictor. The migration rates ...
Tobias Naaf's user avatar
  • 33
0 votes
0 answers
31 views

I am a CS undergrad with some basic stats courses under my belt. I am now taking a time series course, which thus far seems to just apply the same statistical concepts to time-related data, which ...
QuantQuontQuint's user avatar
  • 1
4 votes
1 answer
173 views

I’m modeling body mass in relation to differente variables, but I also need to account for spatial autocorrelation. My current approach uses spaMM with a Matérn correlation term based on coordinates. <...
Leounaa's user avatar
  • 43
0 votes
0 answers
18 views

I have a timeseries of species abundances (continuous variable). I calculated the dissimilarity (Bray Curtis coefficient) between all time steps, and now I want to model this dissimilarities against ...
Rafaela Granzotti's user avatar
  • 1
1 vote
0 answers
25 views

I have a gridded dataset indexed by time and space, represented as a $m \times n$ array. I'm following along with Eq. 10 in this paper to partition the variance in this data over space and time. ...
ganzk's user avatar
  • 11
0 votes
0 answers
43 views

I have a data vector for which I calculated ACF and PACF estimates using R's base acf() and <...
Bogaso's user avatar
  • 1,063
0 votes
0 answers
42 views

I'm in need of some advice on how to account for residual dependence in a fitted GAM that doesn't appear to be driven by temporal structure in the data. In summary, I am working with a long-term ...
Jeremy Hemberger's user avatar
  • 11
0 votes
0 answers
26 views

I am running SDM model with time and spatial fixed effects. When I apply, Moran's I test on the residuals of the model, I get the significant spatial autocorrelation. Why is it the case that I get so ...
Turkana's user avatar
  • 1
0 votes
0 answers
50 views

I have question regarding event studies (pre-event data is observed, an event occurs at $t=e$, then following the treatment is assumed to be in-effect.) There are multiple approaches to event study ...
jbuddy_13's user avatar
  • 3,970
1 vote
0 answers
46 views

I'm deep diving into the ARCH model and i had a doubt. While in AR or ARDL model the autocorrelation is a huge problem and the models themselves are shaped for fixing it, I've been reading that, in ...
Giuseppe's user avatar
  • 11
0 votes
0 answers
93 views

I have a seemingly very simple question that I cannot find the answer to. When performing linear regression, we are assuming that the correlations between residuals is zero. This makes sense to me ...
Joshua Schroijen's user avatar
  • 235
2 votes
0 answers
42 views

I am following Wieling (2018) to fit a gam model. After checking for the autocorrelation of residuals, I added the autocorrelation corrections suggested by Wieling. Although the autocorrelation is ...
user avatar
0 votes
0 answers
70 views

I have monthly time series $\{y_t\}$ and $\{x_t\}$ (continuous variables) with just over 200 observations. I model $y_t$ conditional on $x_t$ them as follows: $$ y_t = \alpha_1Jan_t + \dots + \alpha_{...
Richard Hardy's user avatar
  • 71.5k
5 votes
1 answer
140 views

I am trying to understand species presence (1/0) within protected areas in Africa using a set of predictor variables. My data is not nested and contains no hierarchies within it. My question pertains ...
Tammy's user avatar
  • 91
3 votes
1 answer
385 views

When I have a time series that I want to model as an autoregressive series, I express my value at time t as a linear combination of previous values: $$y_t=\sum_{i=1}^p\theta_iy_{t-i}+e_t$$ Right, but ...
Curious student's user avatar
  • 125
8 votes
1 answer
188 views

I am using XGBoost to forecast electricity prices. In classical time series models such as ARIMA, it is common to evaluate the model by analyzing the residuals using tools like the Ljung-Box test and ...
Marie's user avatar
  • 135
1 vote
1 answer
146 views

This is a follow-up question to my last one (larger dataset found here). I was able to code a variogram (gstat::variogram()), but the resultant plot is a bit ...
Nate's user avatar
  • 2,537
6 votes
1 answer
173 views

I'm not sure if this question is unique to count data, but I'm having trouble finding a means of detecting temporal autocorrelation between fish count observations recorded close in time. Although I ...
Nate's user avatar
  • 2,537
0 votes
0 answers
61 views

I would like to ask regarding my research. I am going to use the cointegration tests Trace and Eigen, but I need optimal lags for them. Then I want to use VAR or VECM model, but it will be based on ...
Katarína's user avatar
  • 1
1 vote
1 answer
76 views

My model: gam(response ~ s(days, k=9) + s(days, by=subject, k=9, m=1) + covariates. I used this approach https://stackoverflow.com/questions/78182559/...
aim6789's user avatar
  • 141
3 votes
2 answers
229 views

Autocorrelation exists and this was tested through Wooldridge test, to account for this I included lags of FDI and HC which corrected this. Is lagging the right way to correct autocorrelation, or is ...
user avatar
1 vote
0 answers
57 views

I have an unbalanced panel (N = 39, T = 14). I aim to investigate the long- and short-run causal relationships between firm-specific variables and firm risk. Based on the findings of non-stationarity ...
Pankaj Kumar Swain's user avatar
  • 11
0 votes
0 answers
35 views

I was looking at my introductory course in econometrics, and found that cross sectional data is generally non autocorrelated. I started to wonder when the 'generally' would not hold. I was thinking ...
wtr8m12's user avatar
  • 11
3 votes
1 answer
86 views

I have over 100 sites where forest carbon was measured once in 2010 and again in 2020. I want to see if the second measurement is significantly dependent on the first one. I think this might also be ...
frandude's user avatar
  • 217
1 vote
0 answers
72 views

Forest carbon has been measured for over 100 independent sites, and each site has been measured at two time points 10 years apart. At observation 1, each site has an amount of carbon that is ...
frandude's user avatar
  • 217
0 votes
0 answers
36 views

Setup: I applied a Gaussian process regression on several inhomogeneous time series. The GP kernels were motivated by a physical understanding of the phenomenon; the posteriors are well-sampled and ...
pedrofigueira's user avatar
  • 859
1 vote
0 answers
57 views

I'm looking at the simulation accuracy of a model that predicts forest carbon. I'm comparing these simulated values against measurements of forest carbon at specific sites. Each site has had forest ...
frandude's user avatar
  • 217
1 vote
1 answer
122 views

When we want to train an ML model on a sample, it's better to have an i.i.d training sample. If the sample is autocorrelated we need much more data to actually train the model. Intuitively this makes ...
ferwfrewgfrewgerccwc's user avatar
  • 31
0 votes
0 answers
57 views

I have a problem that has been plaguing me over the past three days. I have an unbalanced panel. T=52, N=39, obs. = 1565. I am using a fixed effects model with time and country level fixed effects. ...
Trutz's user avatar
  • 11
1 vote
1 answer
118 views

I have clustered binary data collected at three time points, with the following variables: uid: Unit ID (individual unit) cid: ...
schotti's user avatar
  • 600
0 votes
1 answer
312 views

I want to check/ better understand how to model seasonality and autocorrealtion in GAMs. Here's an example of my data. It is count data collected monthly, but I only want the trend at the yearly ...
Justin Murphy's user avatar
  • 61
1 vote
0 answers
60 views

I was reading this post: Analyse ACF and PACF plots, and in the comment it's noted that the data has a left skew which is a problem for AR process. I am wondering why this is a problem and what are ...
autoregressive_monoid's user avatar
  • 11
1 vote
0 answers
43 views

Suppose I have a series $a_1, a_2, a_3, \cdots, a_{qN}$; its standard deviation is called $\sigma_a$. Then define $b_j = \frac{1}{q}\sum_{i=1}^q a_{q(j-1)+i}$ as another series $b_1, b_2, \cdots, b_N$;...
an offer can't refuse's user avatar
  • 143
0 votes
0 answers
64 views

I've constructed a model in mgcv, but it seems there are still patterns in the residuals indicating some strong temporal auto-correlation. The data are irregularly spaced, so I think I need to use the ...
Nate's user avatar
  • 2,537
5 votes
2 answers
932 views

I am trying to understand properties of different time series models. In order to be a white noise $w_t$ must follow three conditions: $E(w_t) = 0$, $Var(w_t) = \sigma^2$, and $cov(w_t, w_s) = 0$, ...
stats_b's user avatar
  • 361
0 votes
0 answers
47 views

I have made a GJR-GARCH(1,1) estimate for various MSCI indices and time periods. There is no autocorrelation (AC) in the squared standardised residuals, but there is some in standardised residuals. I ...
Lea's user avatar
  • 1
3 votes
1 answer
170 views

I am working on a project analyzing olive plantation data, where I aim to simulate the relationship between investment costs (Costs), revenues (...
Barbab's user avatar
  • 1,036
4 votes
2 answers
142 views

I am currently trying to fit a HGAM to model differences in daily activity patterns of fish in two treatments. Data were collected with high-resolution telemetry, and I currently have estimates of ...
Jack B's user avatar
  • 105
0 votes
0 answers
53 views

I have these 1920 observations with shifted distribution. For some reasons I want to use all of observations for time series modeling but I have difficulty in reading the ACF plot. I need detail ...
Aulia Rahman's user avatar
  • 1
3 votes
1 answer
335 views

I was taught the autocorrelation in a time-series at lag $k$ is the correlation between all pairs of values separated by this lag. Suppose I want to give it a go and calculate it manually for lag 1. ...
Mihail's user avatar
  • 582
1 vote
0 answers
72 views

I am trying to determine if I can use the y-intercept of a simple linear regression to predict the slope but I want to confirm whether the y-intercept is spuriously correlated with the slope. Would ...
Annabelle's user avatar
  • 11
1 vote
0 answers
65 views

I am working on determining the Granger Causality of two time series. One thing to note is that for my specific project, I have around 100k time series across two different dimensions, say Product ...
jmoore00's user avatar
  • 391
1 vote
1 answer
64 views

I am reading Econometric Analysis by Greene (7th edition, 2011). On page 350, he discusses the random effects regression model: $$ \begin{aligned} y_{it} &= x'_{it} \beta + E[z'_i \alpha] + \left[ ...
s5s's user avatar
  • 715
2 votes
2 answers
161 views

I am using R to smooth time series using Generalised Additive Models (GAMs). A preceding question concerned uncertain serial autocorrelation in the residuals. I was impressed by the diagnostic plots ...
stweb's user avatar
  • 559
1 vote
1 answer
68 views

I conducted a Breusch-Godfrey (BG) test after performing a regression, and the results showed that the p-value was 0.12, indicating that there is no serial correlation. However, the two-time lagged ...
Xu Yang's user avatar
  • 41
5 votes
1 answer
167 views

I wish to use Generalised Additive Models (GAMs) to smooth count time series and estimate first derivatives, i.e. to identify periods where the counts are increasing, stable or decreasing. I'm using R ...
stweb's user avatar
  • 559
1 vote
0 answers
47 views

I have two questions regarding the degrees of freedom for the Ljung-Box test on residuals in case of different AR(p) models: In case of a model with non-consecutive lags: As I understand it, one has ...
stat1002's user avatar
  • 51
1 vote
0 answers
85 views

I have two sets of data points: Each data point (y_t) in the first set represents the annualized return for the next ten years of monthly return data, i.e., this is forward-looking data, so my last ...
Andrew's user avatar
  • 11

1
2 3 4 5
…
36