dx.gam_allnat: Function to decompose the covariate x into an ALL, an ANT and...
In thaos/FARallnat: FAR analysis on European Summer Temperature

dx.gam_allnat uses the following gam model x_all = beta_nat * nat + s(ant)

dx.raw keeps the simulated EBM forcings as it is

dx.lm_gno uses the follwing linear model : x_all = beta_nat * nat + beta_ghg * ghg + beta_others * others

dx.gam_gno uses the follwing linear model : gam model x_all = beta_nat * nat + beta_ghg * ghg + s(others)

dx.gam_gno fits EBM using ordinary least squares for the decomposition

dx.gam_allnat(bsamples, bindexes, y = "y", x = "x", ant = "ant",
  nat = "nat", time = "time", knots = NULL, fixed = FALSE,
  correct_ant_bias = TRUE, pre_ind = c(1850, 1879),
  reuse_ant_bias = FALSE)

dx.raw(bsamples, bindexes, y = "y", all = "all", ant = "ant",
  nat = "nat", time = "time")

dx.lm_gno(bsamples, bindexes, y = "y", x = "x", ghg = "ghg",
  nat = "nat", others = "others", time = "time",
  correct_ant_bias = TRUE, pre_ind = c(1850, 1879),
  reuse_ant_bias = FALSE)

dx.gam_gno(bsamples, bindexes, y = "y", x = "x", ghg = "ghg",
  nat = "nat", others = "others", time = "time",
  correct_ant_bias = TRUE, pre_ind = c(1850, 1879),
  reuse_ant_bias = FALSE)

dx.ebm_fit(bsamples, bindexes, y = "y", x = "x", time = "time")

`bsamples`	: a list of data.frame with the original dataset merged with the bootstrap EBM simulations
`bindexes`	: a list of vectors of indexes indicating the rows that will be used for the bootstrap
`y`	the name of the variable of interest
`x`	the name of the variable the covariate that will be decomposed
`ant`	the name of the variable that will be used as the ANT variable
`nat`	the name of the variable that will be used as the NAT variable
`time`	the name of the variable that will be used as the time
`knots`	the numbers of knots wanted for the spline in the gam fit
`fixed`	wether the numbers of knots is allowed to vary or not
`correct_ant_bias`	whether the variables that are anthropogenic need to have their mean shifted to zero in the pre-industrial.
`pre_ind`	period of the pre-industrial if the anthropogenic variables need to be shifted.
`reuse_ant_bias`	wheter only the shift computed on the first bootstrap samples is used to reshift the other bootstrap samples. Otherwise each bootstrap samples is correctied by its own shift.
`all`	the name of the variable that will be used as the ALL variable
`ghg`	the name of the variable that will be used as the GHG variable
`others`	the name of the variable that will be used as the OTHERS variable

two list :

bsamples a list of data.frame containing the bootstrap samples with the decomposition of the covariate x into x_all, x_nat and x_ant
osamples a list of data.frame containing the completed/original datasets for each simulation of the EBM as well as the decomposition of the covariate x into x_all, x_nat and x_ant

thaos/FARallnat documentation built on May 25, 2019, 8:18 a.m.