In reyzaguirre/pepa: Package for the Execution of Pre Cooked Analysis

library(knitr)
opts_chunk$set(echo = FALSE, comment = NA)

# Pass arguments

dfr <- params$dfr
vars <- params$vars
factors <- params$factors
rep <- params$rep
maxp <- params$maxp
pe <- params$pe
se <- params$se

# Check factors structure

out <- ck.fs(dfr, factors, rep)
dfr <- out$dfr
nla <- out$nl[1]
nlb <- out$nl[2]
nrep <- out$nrep
nmis.fac <- out$nmis.fac

# Everything as character

dfr[, factors[1]] <- as.character(dfr[, factors[1]])
dfr[, factors[2]] <- as.character(dfr[, factors[2]])

1. Model specification and data description

The data frame has two factors with r nla and r nlb levels. The experimental design is a completely randomized design with r nrep replications. The statistical model is $$ y_{ijk} = \mu + \alpha_i + \beta_j + (\alpha\beta){ij} + \epsilon{ijk} $$ where

$y_{ijk}$ is the observed response with level $i$ of factor A, level $j$ of factor B, and replication $k$.
$\mu$ is the mean response over all levels of factor A, factor B, and replications.
$\alpha_i$ is the effect for level $i$ of factor A.
$\beta_j$ is the effect for level $j$ of factor B.
$(\alpha\beta)_{ij}$ is the interaction effect between level $i$ of factor A and level $j$ of factor B.
$\epsilon_{ijk}$ is the error term.

In this model we assume that the errors are independent and have a normal distribution with common variance, that is, $\epsilon_{ijk} \sim N(0,\sigma_{\epsilon}^2)$.

r if (nmis.fac == 1) paste("Note: There is", nmis.fac, "row with missing values for classifications factors. This row has been deleted.") r if (nmis.fac > 1) paste("Note: There are", nmis.fac, "rows with missing values for classifications factors. These rows have been deleted.")

out <- NULL
for (i in 1:length(vars)) {
  lc <- ck.f(dfr, vars[i], factors, rep)
  if (lc$nt.0 == 0 & lc$nrep > 1 & lc$nt.mult == 0 & lc$pmis <= maxp) {
    out <- c(out, knit_expand('child_f.Rmd'))
  } else {
    out <- c(out, knit_expand('child_f_fail.Rmd'))
  }
}