Articles by GAMbler

State-Space Vector Autoregressions in mvgam

September 5, 2024 | GAMbler

Vector Autoregressions (VAR models), also known in the ecological literature as Multivariate Autoregressions (MAR models), are a class of Dynamic Linear Models that offer a principled way to model delayed and contemporaneous interactions among sets of multiple time series. These models are widely used in econometrics and psychology, among other ...

Incorporating time-varying seasonality in forecast models

June 10, 2024 | GAMbler

Seasonality is very common in real-world time series. Many series vary in periodic, regular ways. For example, ice cream sales tend to be higher in warmer holiday months, while counts of migratory birds fluctuate strongly around the annual migration cycle. Because of how pervasive seasonality is, many time series and ...

First release of mvgam(v1.1.0) to CRAN

May 7, 2024 | GAMbler

Welcoming mvgam to the Comprehensive R Archive Network (CRAN) The goal of mvgam is to use a Bayesian framework to estimate parameters of Dynamic Generalized Additive Models (DGAMs) for time series with dynamic trend component... [Read more...]

How to interpret and report nonlinear effects from Generalized Additive Models

March 20, 2024 | GAMbler

What are Generalized Additive Models (GAMs)? Generalized Additive Models (GAMs) are flexible tools that replace one or more predictors in a Generalized Linear Model (GLM) with smooth functions of predictors. These are helpful for learning arbitrarily complex, nonlinear relationships between predictors and conditional responses without needing a priori expectations about ... [Read more...]

Simulating data for model exploration in R

March 11, 2024 | GAMbler

Purpose The following passage is quoted directly from Regression and other stories (Gelman, Hill and Vehtari 2022). “Simulation of random variables is important in applied statistics for several reasons. First, we use p...

Using Stan for logistic regressions with detection error

March 9, 2024 | GAMbler

Required libraries cmdstanr Purpose and model introduction This script simulates binary observations of an imperfectly observed data generating process (i.e. our measurements are made with error). I...

Phylogenetic smoothing using mgcv

February 23, 2024 | GAMbler

I have been highly interested in multivariate time series modeling for the last few years and have spent a lot of time working out how to use hierarchical GAMs to tackle many of the questions I’m interested in. As you may already know, GAMs affor...

Using Stan to model geostatistical count data with distance matrices

November 15, 2023 | GAMbler

Required libraries cmdstanr MASS raster ggplot2 viridis Purpose and model introduction This script simulates Poisson observations of an underlying spatially autocorrelated data generating process. I...

Distributed lags (and hierarchical distributed lags) using mgcv and mvgam

August 14, 2023 | GAMbler

Here we will use mgcv to estimate parameters of nonlinear distributed lag models. These models are used to describe simultaneously non-linear and delayed functional relationships between a covariate and a response, and are sometimes referred to as exposure-lag-response models. If we assume \(\tilde{\boldsymbol{y}}_{t}\) is the conditional expectation ...

Temporal autocorrelation in GAMs and the mvgam package

July 15, 2023 | GAMbler

Generalized Additive Models (GAMs) are flexible tools that have found particular application in the analysis of time series data. In ecology, a host of recent papers and workshops have drawn attention to the power of GAMs for addressing complex ecological analyses. For example, Gavin Simpson regularly hosts workshops on GAMs ... [Read more...]

Copyright © 2025 | MH Corporate basic by MH Themes