Someone asked me the other day about whether glmmfields can fit 1-D spatial models (like along a coastline). The answer is yes -- it's maybe not the fastest implementation of 1-D gaussian process models, but we can do that by setting latitude or longitude to some constant. For example,
Someone asked me the other day about whether glmmfields can fit 1-D spatial models (like along a coastline). The answer is yes -- it's maybe not the fastest implementation of 1-D gaussian process models, but we can do that by setting latitude or longitude to some constant. For example,
s <- sim_glmmfields(n_draws = 12, n_knots = 12, gp_theta = 1.5, gp_sigma = 0.2, sd_obs = 0.2)
s$dat$lon = 0 # set to arbitrary constant
m <- glmmfields(y ~ 0, time = "time", lat = "lat", lon = "lon", data = s$dat, nknots = 12, iter = 1000, chains = 4, seed = 1)