lm_optim: Optimize parameters for onlineforecast model fitted with LM

Description

Optimize parameters (transformation stage) of LM model

Usage

lm_optim(
  model,
  data,
  kseq = NA,
  scorefun = rmse,
  cachedir = "",
  cachererun = FALSE,
  printout = TRUE,
  method = "L-BFGS-B",
  ...
)

Value

Result object of optim(). Parameters resulting from the optimization can be found from result$par

Arguments

model: The onlineforecast model, including inputs, output, kseq, p
data: The data.list including the variables used in the model.
kseq: The horizons to fit for (if not set, then model$kseq is used)
scorefun: The function to be score used for calculating the score to be optimized.
cachedir: A character specifying the path (and prefix) of the cache file name. If set to "", then no cache will be loaded or written. See https://onlineforecasting.org/vignettes/nice-tricks.html for examples.
cachererun: A logical controlling whether to run the optimization even if the cache exists.
printout: A logical determining if the score function is printed out in each iteration of the optimization.
method: The method argument for optim.
...: Additional parameters to optim

Details

This is a wrapper for optim to enable easy use of bounds and caching in the optimization.

Examples

Run this code


# Take data
D <- subset(Dbuilding, c("2010-12-15", "2011-01-01"))
D$y <- D$heatload
# Define a simple model 
model <- forecastmodel$new()
model$add_inputs(Ta = "lp(Ta, a1=0.9)",
                 mu = "one()")
# Before fitting the model, define which points to include in the evaluation of the score function
D$scoreperiod <- in_range("2010-12-20", D$t)
# And the sequence of horizons to fit for
model$kseq <- 1:6

# Now we can fit the model and get the score, as it is
lm_fit(model=model, data=D, scorefun=rmse, returnanalysis=FALSE)
# Or we can change the low-pass filter coefficient
lm_fit(c(Ta__a1=0.99), model, D, rmse, returnanalysis=FALSE)

# This could be passed to optim() (or any optimizer).
# See \code{forecastmodel$insert_prm()} for more details.
optim(c(Ta__a1=0.98), lm_fit, model=model, data=D, scorefun=rmse, returnanalysis=FALSE,
      lower=c(Ta__a1=0.4), upper=c(Ta__a1=0.999), method="L-BFGS-B")

# lm_optim is simply a helper it makes using bounds easiere and enables caching of the results
# First add bounds for lambda (lower, init, upper)
model$add_prmbounds(Ta__a1 = c(0.4, 0.98, 0.999))

# Now the same optimization as above can be done by
val <- lm_optim(model, D)
val