The model that I have arrived at is a zero-inflated generalized linear mixed-effects model (ZIGLMM). Several packages that I have attempted to use to fit such a model include glmmTMB and glmmADMB in R. My question is: is it possible to account for spatial autocorrelation using such a model and if so, how can it be done?Linear mixed models allow for modeling fixed, random and repeated effects in analysis of variance models. “Factor effects are either fixed or random depending on how levels of factors that appear in the study are selected. An effect is called fixed if the levels in the study represent all possible levels of theDec 24, 2014 · Is it accurate to say that we used a linear mixed model to account for missing data (i.e. non-response; technology issues) and participant-level effects (i.e. how frequently each participant used ... Ultimately I'd like to include spatial autocorrelation with corSpatial(form = ~ lat + long) in the GAMM model, or s(lat,long) in the GAM model, but even in basic form I can't get the model to run. If it helps understand the structure of the data, I've added dummy code below (with 200,000 rows):Sep 16, 2018 · Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ... What is autocorrelation? Generalized Additive Mixed Effects Models have several components: Smooth terms for covariates; Random Effects: Intercepts, Slopes and Smooths. Categorical Predictors; Interactions of (1)-(3) We can add one more component for autocorrelation: modeling the residuals: Covariance structure for the residuals. Sep 16, 2018 · Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ... Apr 15, 2021 · Yes. How can glmmTMB tell how far apart moments in time are if the time sequence must be provided as a factor? The assumption is that successive levels of the factor are one time step apart (the ar1 () covariance structure does not allow for unevenly spaced time steps: for that you need the ou () covariance structure, for which you need to use ... Yes. How can glmmTMB tell how far apart moments in time are if the time sequence must be provided as a factor? The assumption is that successive levels of the factor are one time step apart (the ar1 () covariance structure does not allow for unevenly spaced time steps: for that you need the ou () covariance structure, for which you need to use ...Spatial and temporal autocorrelation can be problematic because they violate the assumption that the residuals in regression are independent, which causes estimated standard errors of parameters to be biased and causes parametric statistics no longer follow their expected distributions (i.e. p-values are too low).Subject. Re: st: mixed effect model and autocorrelation. Date. Sat, 13 Oct 2007 12:00:33 +0200. Panel commands in Stata (note: only "S" capitalized!) usually accept unbalanced panels as input. -glamm- (remember the dashes!), which you can download from ssc (by typing: -ssc install gllamm-), allow for the option cluster, which at least partially ...It is evident that the classical bootstrap methods developed for simple linear models should be modified to take into account the characteristics of mixed-effects models (Das and Krishen 1999). In ...(1) this assumes the temporal pattern is the same across subjects; (2) because gamm() uses lme rather than lmer under the hood you have to specify the random effect as a separate argument. (You could also use the gamm4 package, which uses lmer under the hood.) You might want to allow for temporal autocorrelation. For example,Mixed Models (GLMM), and as our random effects logistic regression model is a special case of that model it fits our needs. An overview about the macro and the theory behind is given in Chapter 11 of Littell et al., 1996. Briefly, the estimating algorithm uses the principle of quasi-likelihood and an approximation to the likelihood function of ... PROC MIXED in the SAS System provides a very flexible modeling environment for handling a variety of repeated measures problems. Random effects can be used to build hierarchical models correlating measurements made on the same level of a random factor, including subject-specific regression models, while a variety of covariance and Is it accurate to say that we used a linear mixed model to account for missing data (i.e. non-response; technology issues) and participant-level effects (i.e. how frequently each participant used ...of freedom obtained by the same method used in the most recently fit mixed model. If option dfmethod() is not specified in the previous mixed command, option small is not allowed. For certain methods, the degrees of freedom for some linear combinations may not be available. See Small-sample inference for fixed effects in[ME] mixed for more ... I have temporal blocks in my data frame, so I took the effect of time dependency through a random intercept in a glmer model. Now I want to test the spatial autocorrelation in the residuals but I’m not sure if the test procedure based on the residual is the same as for the fixed-effect models since now I have time dependency.In the present article, we suggested an extension of the mixed-effects location scale model that allows a researcher to include random effects for the means, the within-person residual variance, and the autocorrelation.The first model was a longitudinal mixed-effect model with a first-order autocorrelation structure, and the second model was the E-MELS. Both were implemented as described above. The third model was a longitudinal mixed-effect model with a Lasso penalty. Random intercept + Autocorrelation structure on the errors, and; Autocorrelation structure on the errors only (using gls() command). I fit model 3 because I've been taught that sometimes an autocorrelation structure is enough for longitudinal data. For model 1, variance of random effect (intercept) was 676.9 (and accounted for 62% of total ...the mixed-effect model with a first-order autocorrelation structure. The model was estimated using the R package nlme and the lme function (Pinheiro et al., 2020 ).Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ...Linear Mixed Effects Models. Linear Mixed Effects models are used for regression analyses involving dependent data. Such data arise when working with longitudinal and other study designs in which multiple observations are made on each subject. Some specific linear mixed effects models are. Random intercepts models, where all responses in a ...Mixed Effects Models - Autocorrelation. Jul. 1, 2021 • 0 likes • 171 views. Download Now. Download to read offline. Education. Lecture 19 from my mixed-effects modeling course: Autocorrelation in longitudinal and time-series data. Scott Fraundorf Follow.Is it accurate to say that we used a linear mixed model to account for missing data (i.e. non-response; technology issues) and participant-level effects (i.e. how frequently each participant used ...Eight models were estimated in which subjects nervousness values were regressed on all aforementioned predictors. The first model was a standard mixed-effects model with random effects for the intercept and the slope but no autocorrelation (Model 1 in Tables 2 and 3). The second model included such an autocorrelation (Model 2).Eight models were estimated in which subjects nervousness values were regressed on all aforementioned predictors. The first model was a standard mixed-effects model with random effects for the intercept and the slope but no autocorrelation (Model 1 in Tables 2 and 3). The second model included such an autocorrelation (Model 2). Segmented linear regression models are often fitted to ITS data using a range of estimation methods [8,9,10,11]. Commonly ordinary least squares (OLS) is used to estimate the model parameters ; however, the method does not account for autocorrelation. Other statistical methods are available that attempt to account for autocorrelation in ...Subject. Re: st: mixed effect model and autocorrelation. Date. Sat, 13 Oct 2007 12:00:33 +0200. Panel commands in Stata (note: only "S" capitalized!) usually accept unbalanced panels as input. -glamm- (remember the dashes!), which you can download from ssc (by typing: -ssc install gllamm-), allow for the option cluster, which at least partially ... Feb 10, 2022 · An extension of the mixed-effects growth model that considers between-person differences in the within-subject variance and the autocorrelation. Stat Med. 2022 Feb 10;41 (3):471-482. doi: 10.1002/sim.9280. A random effects model that contains only random intercepts, which is the most common use of mixed effect modeling in randomized trials, assumes that the responses within subject are exchangeable. This can be seen from the statement of the linear mixed effects model with random intercepts.However, in the nlme R code, both methods inhabit the ‘correlation = CorStruc’ code which can only be used once in a model. Therefore, it appears that either only spatial autocorrelation or only temporal autocorrelation can be addressed, but not both (see example code below).Ultimately I'd like to include spatial autocorrelation with corSpatial(form = ~ lat + long) in the GAMM model, or s(lat,long) in the GAM model, but even in basic form I can't get the model to run. If it helps understand the structure of the data, I've added dummy code below (with 200,000 rows):spaMM fits mixed-effect models and allow the inclusion of spatial effect in different forms (Matern, Interpolated Markov Random Fields, CAR / AR1) but also provide interesting other features such as non-gaussian random effects or autocorrelated random coefficient (ie group-specific spatial dependency). spaMM uses a syntax close to the one used ...Mixed-effects models allow multiple levels of variability; AKA hierarchical models, multilevel models, multistratum models; Good references on mixed-effects models: Bolker [1–3] Gelman & Hill [4] Pinheiro & Bates [5].A comparison to mixed models. We noted previously that there were ties between generalized additive and mixed models. Aside from the identical matrix representation noted in the technical section, one of the key ideas is that the penalty parameter for the smooth coefficients reflects the ratio of the residual variance to the variance components for the random effects (see Fahrmeier et al ... 3. MIXED EFFECTS MODELS 3.1 Overview of mixed effects models When a regression contains both random and fixed effects, it is said to be a mixed effects model, or simply, a mixed model. Fixed effects are those with which most researchers are familiar. Any covariate that is assumed to have the same effect for all responses throughout theSep 16, 2018 · Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ... Abstract. The ‘DHARMa’ package uses a simulation-based approach to create readily interpretable scaled (quantile) residuals for fitted (generalized) linear mixed models. Currently supported are linear and generalized linear (mixed) models from ‘lme4’ (classes ‘lmerMod’, ‘glmerMod’), ‘glmmTMB’, ‘GLMMadaptive’ and ‘spaMM ...a random effect for the autocorrelation. After introducing the extended mixed-effect location scale (E-MELS), ... mixed-effect models that have been, for example, combined with Lasso regression (e ...The nlme package allows you to fit mixed effects models. So does lme4 - which is in some ways faster and more modern, but does NOT model heteroskedasticity or (!spoiler alert!) autocorrelation. Let’s try a model that looks just like our best model above, but rather than have a unique Time slope My approach is to incorporate routes and year as random effects in generalized mixed effects models as shown below (using lme4 package). But, I am not sure how well autocorrelation is modeled adequately in this way. glmer (Abundance ~ Area_harvested + (1 | route) + (1 | Year), data = mydata, family = poisson) Although I specified Poisson above ...Mixed Effects Models - Autocorrelation. Jul. 1, 2021 • 0 likes • 171 views. Download Now. Download to read offline. Education. Lecture 19 from my mixed-effects modeling course: Autocorrelation in longitudinal and time-series data. Scott Fraundorf Follow.Yes. How can glmmTMB tell how far apart moments in time are if the time sequence must be provided as a factor? The assumption is that successive levels of the factor are one time step apart (the ar1 () covariance structure does not allow for unevenly spaced time steps: for that you need the ou () covariance structure, for which you need to use ...Linear mixed model fit by maximum likelihood [’lmerMod’] AIC BIC logLik deviance df.resid 22.5 25.5 -8.3 16.5 17 Random effects: Groups Name Variance Std.Dev. operator (Intercept) 0.04575 0.2139 *** Operator var Residual 0.10625 0.3260 estimate is smaller. Number of obs: 20, groups: operator, 4 Results in smaller SE for the overall Fixed ... Subject. Re: st: mixed effect model and autocorrelation. Date. Sat, 13 Oct 2007 12:00:33 +0200. Panel commands in Stata (note: only "S" capitalized!) usually accept unbalanced panels as input. -glamm- (remember the dashes!), which you can download from ssc (by typing: -ssc install gllamm-), allow for the option cluster, which at least partially ...in nlme, it is possible to specify the variance-covariance matrix for the random effects (e.g. an AR (1)); it is not possible in lme4. Now, lme4 can easily handle very huge number of random effects (hence, number of individuals in a given study) thanks to its C part and the use of sparse matrices. The nlme package has somewhat been superseded ... Abstract. The ‘DHARMa’ package uses a simulation-based approach to create readily interpretable scaled (quantile) residuals for fitted (generalized) linear mixed models. Currently supported are linear and generalized linear (mixed) models from ‘lme4’ (classes ‘lmerMod’, ‘glmerMod’), ‘glmmTMB’, ‘GLMMadaptive’ and ‘spaMM ...Generalized additive models were flrst proposed by Hastie and Tibshirani (1986, 1990). These models assume that the mean of the response variable depends on an additive pre-dictor through a link function. Like generalized linear models (GLMs), generalized additive models permit the response probability distribution to be any member of the ...I'm trying to model the evolution in time of one weed species (E. crus galli) within 4 different cropping systems (=treatment). I have 5 years of data spaced out equally in time and two repetitions (block) for each cropping system. Hence, block is a random factor. Measures were repeated each year on the same block (--> repeated measure mixed ...Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ...May 22, 2018 · 10.8k 7 39 67. 1. All LMMs correspond to a multivariate normal model (while the converse is not true) with a structured variance covariance matrix, so "all" you have to do is to work out the marginal variance covariance matrix for the nested random-effect model and fit that - whether gls is then able to parameterize that model is then the next ... The following simulates and fits a model where the linear predictor in the logistic regression follows a zero-mean AR(1) process, see the glmmTMB package vignette for more details.We conducted a small simulation study to investigate whether an extension of the mixed-effect model that considers between-person differences in the Level 1 variance and the autocorrelation (i.e., the E-MELS) yields more precise forecasts than a standard longitudinal mixed-effect model.To do this, you would specify: m2 <- lmer (Obs ~ Day + Treatment + Day:Treatment + (Day | Subject), mydata) In this model: The intercept if the predicted score for the treatment reference category at Day=0. The coefficient for Day is the predicted change over time for each 1-unit increase in days for the treatment reference category. An extension of the mixed-effects growth model that considers between-person differences in the within-subject variance and the autocorrelation. Stat Med. 2022 Feb 10;41 (3):471-482. doi: 10.1002/sim.9280.Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ...Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ...Linear mixed model fit by maximum likelihood [’lmerMod’] AIC BIC logLik deviance df.resid 22.5 25.5 -8.3 16.5 17 Random effects: Groups Name Variance Std.Dev. operator (Intercept) 0.04575 0.2139 *** Operator var Residual 0.10625 0.3260 estimate is smaller. Number of obs: 20, groups: operator, 4 Results in smaller SE for the overall Fixed ... 3. MIXED EFFECTS MODELS 3.1 Overview of mixed effects models When a regression contains both random and fixed effects, it is said to be a mixed effects model, or simply, a mixed model. Fixed effects are those with which most researchers are familiar. Any covariate that is assumed to have the same effect for all responses throughout theSep 16, 2018 · Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ... of freedom obtained by the same method used in the most recently fit mixed model. If option dfmethod() is not specified in the previous mixed command, option small is not allowed. For certain methods, the degrees of freedom for some linear combinations may not be available. See Small-sample inference for fixed effects in[ME] mixed for more ... How is it possible that the model fits perfectly the data while the fixed effect is far from overfitting ? Is it normal that including the temporal autocorrelation process gives such R² and almost a perfect fit ? (largely due to the random part, fixed part often explains a small part of the variance in my data). Is the model still interpretable ?Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ...To use such data for predicting feelings, beliefs, and behavior, recent methodological work suggested combinations of the longitudinal mixed-effect model with Lasso regression or with regressi … A Lasso and a Regression Tree Mixed-Effect Model with Random Effects for the Level, the Residual Variance, and the Autocorrelation 7. I want to specify different random effects in a model using nlme::lme (data at the bottom). The random effects are: 1) intercept and position varies over subject; 2) intercept varies over comparison. This is straightforward using lme4::lmer: lmer (rating ~ 1 + position + (1 + position | subject) + (1 | comparison), data=d) > ...10.8k 7 39 67. 1. All LMMs correspond to a multivariate normal model (while the converse is not true) with a structured variance covariance matrix, so "all" you have to do is to work out the marginal variance covariance matrix for the nested random-effect model and fit that - whether gls is then able to parameterize that model is then the next ...The “random effects model” (also known as the mixed effects model) is used when the analysis must account for both fixed and random effects in the model. This occurs when data for a subject are independent observations following a linear model or GLM, but the regression coefficients vary from person to person. Infant growth is aEight models were estimated in which subjects nervousness values were regressed on all aforementioned predictors. The first model was a standard mixed-effects model with random effects for the intercept and the slope but no autocorrelation (Model 1 in Tables 2 and 3). The second model included such an autocorrelation (Model 2).Abstract. The use of linear mixed effects models (LMMs) is increasingly common in the analysis of biological data. Whilst LMMs offer a flexible approach to modelling a broad range of data types, ecological data are often complex and require complex model structures, and the fitting and interpretation of such models is not always straightforward.GLMMs. In principle, we simply define some kind of correlation structure on the random-effects variance-covariance matrix of the latent variables; there is not a particularly strong distinction between a correlation structure on the observation-level random effects and one on some other grouping structure (e.g., if there were a random effect of year (with multiple measurements within each year ...Eight models were estimated in which subjects nervousness values were regressed on all aforementioned predictors. The first model was a standard mixed-effects model with random effects for the intercept and the slope but no autocorrelation (Model 1 in Tables 2 and 3). The second model included such an autocorrelation (Model 2). The first model was a longitudinal mixed-effect model with a first-order autocorrelation structure, and the second model was the E-MELS. Both were implemented as described above. The third model was a longitudinal mixed-effect model with a Lasso penalty.1 Answer. Mixed models are often a good choice when you have repeated measures, such as here, within whales. lme from the nlme package can fit mixed models and also handle autocorrelation based on a AR (1) process, where values of X X at t − 1 t − 1 determine the values of X X at t t.Apr 15, 2021 · Yes. How can glmmTMB tell how far apart moments in time are if the time sequence must be provided as a factor? The assumption is that successive levels of the factor are one time step apart (the ar1 () covariance structure does not allow for unevenly spaced time steps: for that you need the ou () covariance structure, for which you need to use ... Mixed Models (GLMM), and as our random effects logistic regression model is a special case of that model it fits our needs. An overview about the macro and the theory behind is given in Chapter 11 of Littell et al., 1996. Briefly, the estimating algorithm uses the principle of quasi-likelihood and an approximation to the likelihood function of ...7. I want to specify different random effects in a model using nlme::lme (data at the bottom). The random effects are: 1) intercept and position varies over subject; 2) intercept varies over comparison. This is straightforward using lme4::lmer: lmer (rating ~ 1 + position + (1 + position | subject) + (1 | comparison), data=d) > ...I am seeking advice on how to effectively eliminate autocorrelation from a linear mixed model. My experimental design and explanation of fixed and random factors can be found here from an earlier question I asked: Crossed fixed effects model specification including nesting and repeated measures using glmm in RTo use such data for predicting feelings, beliefs, and behavior, recent methodological work suggested combinations of the longitudinal mixed-effect model with Lasso regression or with regressi … A Lasso and a Regression Tree Mixed-Effect Model with Random Effects for the Level, the Residual Variance, and the Autocorrelation The “random effects model” (also known as the mixed effects model) is used when the analysis must account for both fixed and random effects in the model. This occurs when data for a subject are independent observations following a linear model or GLM, but the regression coefficients vary from person to person. Infant growth is a Mar 15, 2022 · A random effects model that contains only random intercepts, which is the most common use of mixed effect modeling in randomized trials, assumes that the responses within subject are exchangeable. This can be seen from the statement of the linear mixed effects model with random intercepts.
Gamma mixed effects models using the Gamma() or Gamma.fam() family object. Linear mixed effects models with right and left censored data using the censored.normal() family object. Users may also specify their own log-density function for the repeated measurements response variable, and the internal algorithms will take care of the optimization. . Used toyota tacoma for sale under dollar5000
Dec 11, 2017 · Mixed-effect linear models. Whereas the classic linear model with n observational units and p predictors has the vectorized form. where and are design matrices that jointly represent the set of predictors. Random effects models include only an intercept as the fixed effect and a defined set of random effects. Dec 24, 2014 · Is it accurate to say that we used a linear mixed model to account for missing data (i.e. non-response; technology issues) and participant-level effects (i.e. how frequently each participant used ... The following simulates and fits a model where the linear predictor in the logistic regression follows a zero-mean AR(1) process, see the glmmTMB package vignette for more details.Nov 1, 2019 · Therefore, even greater sampling rates will be required when autocorrelation is present to meet the levels prescribed by analyses of the power and precision when estimating individual variation using mixed effect models (e.g., Wolak et al. 2012; Dingemanse and Dochtermann 2013) Apr 15, 2021 · Yes. How can glmmTMB tell how far apart moments in time are if the time sequence must be provided as a factor? The assumption is that successive levels of the factor are one time step apart (the ar1 () covariance structure does not allow for unevenly spaced time steps: for that you need the ou () covariance structure, for which you need to use ... Jul 25, 2020 · How is it possible that the model fits perfectly the data while the fixed effect is far from overfitting ? Is it normal that including the temporal autocorrelation process gives such R² and almost a perfect fit ? (largely due to the random part, fixed part often explains a small part of the variance in my data). Is the model still interpretable ? At this point, it is important to highlight how spatial data is internally stored in a SpatialGridDataFrame and the latent effects described in Table 7.1. For some models, INLA considers data sorted by column, i.e., a vector with the first column of the grid from top to bottom, followed by the second column and so on. Feb 10, 2022 · An extension of the mixed-effects growth model that considers between-person differences in the within-subject variance and the autocorrelation. Stat Med. 2022 Feb 10;41 (3):471-482. doi: 10.1002/sim.9280. The advantage of mixed effects models is that you can also account for non-independence among "slopes". As you said, you may assume more similarity from fish within tanks, but - e.g. - over time ... Subject. Re: st: mixed effect model and autocorrelation. Date. Sat, 13 Oct 2007 12:00:33 +0200. Panel commands in Stata (note: only "S" capitalized!) usually accept unbalanced panels as input. -glamm- (remember the dashes!), which you can download from ssc (by typing: -ssc install gllamm-), allow for the option cluster, which at least partially ...Apr 15, 2021 · Yes. How can glmmTMB tell how far apart moments in time are if the time sequence must be provided as a factor? The assumption is that successive levels of the factor are one time step apart (the ar1 () covariance structure does not allow for unevenly spaced time steps: for that you need the ou () covariance structure, for which you need to use ... Sep 16, 2018 · Recently I have made good use of Matlab's built-in functions for making linear mixed effects. Currently I am trying to model time-series data (neuronal activity) from cognitive experiments with the fitlme() function using two continuous fixed effects (linear speed and acceleration) and several, hierarchically nested categorical random factors (subject identity, experimental session and binned ... Subject. Re: st: mixed effect model and autocorrelation. Date. Sat, 13 Oct 2007 12:00:33 +0200. Panel commands in Stata (note: only "S" capitalized!) usually accept unbalanced panels as input. -glamm- (remember the dashes!), which you can download from ssc (by typing: -ssc install gllamm-), allow for the option cluster, which at least partially ...A comparison to mixed models. We noted previously that there were ties between generalized additive and mixed models. Aside from the identical matrix representation noted in the technical section, one of the key ideas is that the penalty parameter for the smooth coefficients reflects the ratio of the residual variance to the variance components for the random effects (see Fahrmeier et al ... Jul 7, 2020 · 1 Answer. Mixed models are often a good choice when you have repeated measures, such as here, within whales. lme from the nlme package can fit mixed models and also handle autocorrelation based on a AR (1) process, where values of X X at t − 1 t − 1 determine the values of X X at t t. .