===========================================================
.___ __ __
_________________ __ __ __| _/|__|/ |_
/ ___\_` __ \__ \ | | \/ __ | | \\_ __\
/ /_/ > | \// __ \| | / /_/ | | || |
\___ /|__| (____ /____/\____ | |__||__|
/_____/ \/ \/
grep rough audit - static analysis tool
v2.8 written by @Wireghoul
=================================[justanotherhacker.com]===
r-bioc-multtest-2.46.0/R/EBzzz.R-59- cat(paste("type I error rate =",ifelse(is.null(call.list$typeone),"fwer",call.list$typeone),"\n"))
r-bioc-multtest-2.46.0/R/EBzzz.R:60: nominal<-eval(call.list$alpha)
r-bioc-multtest-2.46.0/R/EBzzz.R:61: if(is.null(eval(call.list$alpha))) nominal<-0.05
r-bioc-multtest-2.46.0/R/EBzzz.R-62- cat("nominal level alpha = ")
##############################################
r-bioc-multtest-2.46.0/R/EBzzz.R-175- cat(paste("prior: ",ifelse(is.null(call.list$prior),"conservative",call.list$prior),"\n\n"))
r-bioc-multtest-2.46.0/R/EBzzz.R:176: nominal<-eval(call.list$alpha)
r-bioc-multtest-2.46.0/R/EBzzz.R-177- if(is.null(nominal)) nominal<-0.05
##############################################
r-bioc-multtest-2.46.0/R/zzz.R-134- points(1:top,x@estimate[topp],pch="o")
r-bioc-multtest-2.46.0/R/zzz.R:135: nominal<-eval(call.list$alpha)
r-bioc-multtest-2.46.0/R/zzz.R-136- if(is.null(nominal)) nominal<-0.05
##############################################
r-bioc-multtest-2.46.0/R/zzz.R-150- points(1:top,stats[topp],pch="o")
r-bioc-multtest-2.46.0/R/zzz.R:151: nominal<-eval(call.list$alpha)
r-bioc-multtest-2.46.0/R/zzz.R-152- if(is.null(nominal)) nominal<-0.05
##############################################
r-bioc-multtest-2.46.0/R/zzz.R-173- cat(paste("Type I error rate: ",err,"\n\n"))
r-bioc-multtest-2.46.0/R/zzz.R:174: nominal<-eval(call.list$alpha)
r-bioc-multtest-2.46.0/R/zzz.R-175- if(is.null(nominal)) nominal<-0.05
##############################################
r-bioc-multtest-2.46.0/R/zzz.R-564- cat(paste("type I error rate =",ifelse(is.null(call.list$typeone),"fwer",call.list$typeone),"\n"))
r-bioc-multtest-2.46.0/R/zzz.R:565: nominal<-eval(call.list$alpha)
r-bioc-multtest-2.46.0/R/zzz.R:566: if(is.null(eval(call.list$alpha))) nominal<-0.05
r-bioc-multtest-2.46.0/R/zzz.R-567- cat("nominal level alpha = ")
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-159-
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:160:In many testing problems, the submodels concern {\em parameters}, i.e., functions of the data generating distribution $P$, $\Psi(P) = \psi= (\psi(m):m=1,\ldots,M)$, such as means, differences in means, correlations, and parameters in linear models, generalized linear models, survival models, time-series models, dose-response models, etc. One distinguishes between two types of testing problems: {\em one-sided tests}, where $H_0(m) = \mathrm{I}(\psi(m) \leq \psi_0(m))$, and {\em two-sided tests}, where $H_0(m) = \mathrm{I}(\psi(m) =
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-161-\psi_0(m))$.
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-198-\end{equation}
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:199:where $n_k$, $\bar{X}_{k,n_k}(m)$, and $\sigma_{k,n_k}^2(m)$ denote, respectively, the sample size, sample means, and sample variances, for patients with test status $k$, $k=0,\, 1$. The null hypotheses are rejected, i.e., the corresponding genes are declared differentially expressed, for large values of the test statistics $T_n(m)$.\\
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-200-
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-207-\end{equation}
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:208:where ${\cal C}_n(m)={\cal C}(T_n,Q_{0n},\alpha)(m)$, $m=1,\ldots,M$, denote possibly random rejection regions. The long notation ${\cal R}(T_n, Q_{0n},\alpha)$ and ${\cal C}(T_n, Q_{0n},\alpha)(m)$ emphasizes that the MTP depends on:
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-209-(i) the {\em data}, $X_1, \ldots, X_n$,
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-375-For a broad class of testing problems, such as the test of single-parameter null hypotheses using $t$-statistics (as in Equation (\ref{anal:mult:e:tstat})), the null distribution $Q_0$ is an $M$--variate Gaussian distribution with mean vector zero and covariance matrix $\Sigma^*(P)$: $Q_0 = Q_0(P) \equiv N(0,\Sigma^*(P))$.
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:376:For tests of means, where the parameter of interest is the $M$--dimensional mean vector $\Psi(P) = \psi = E[X]$, the estimator $\psi_n$ is simply the $M$--vector of sample averages and $\Sigma^*(P)$ is the correlation matrix of $X \sim P$, $Cor[X]$. More generally, for an asymptotically linear estimator $\psi_n$, $\Sigma^*(P)$ is the correlation matrix of the vector influence curve (IC).
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-377-
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-458-\end{equation}
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:459:where $Z^{\circ}(m)$ denotes the $m$th ordered component of $Z = (Z(m): m=1,\ldots,M) \sim Q_0$, so that $Z^{\circ}(1) \geq \ldots \geq Z^{\circ}(M)$.
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-460-For FWER control ($k=0$), the procedure reduces to the {\em single-step maxT procedure}, based on the {\em maximum test statistic}, $Z^{\circ}(1)$.\\
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-463-{\bf Single-step common-quantile procedure.} The set of rejected hypotheses for the {\em $\theta$--controlling single-step common-quantile procedure} is of the form
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:464:${\cal R}_n \equiv \{m: T_n(m)> c_0(m) \}$, where $c_0(m) = Q_{0,m}^{-1}(\delta_0)$ is the $\delta_0$--quantile of the marginal null distribution $Q_{0,m}$ of the $m$th test statistic, i.e., the smallest value $c$ such that $Q_{0,m}(c) = Pr_{Q_0}(Z(m) \leq c) \geq \delta_0$ for $Z \sim Q_0$. Here, $\delta_0$ is chosen as the {\em smallest} (i.e., least conservative) value for which $\theta(F_{R_0}) \leq \alpha$.
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-465-
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-470-\end{equation}
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:471:where $P_0^{\circ}(m)$ denotes the $m$th ordered component of the $M$--vector of unadjusted $p$-values $(P_0(m): m=1,\ldots,M)$, so that $P_0^{\circ}(1) \leq \ldots \leq P_0^{\circ}(M)$.
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-472-For FWER control ($k=0$), one recovers the {\em single-step minP procedure}, based on the {\em minimum unadjusted $p$-value}, $P_0^{\circ}(1)$.
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-494-\end{equation}
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:495:where $Z=(Z(m): m=1,\ldots, M) \sim Q_0$.
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-496-Taking maxima of the probabilities over $h \in \{1, \ldots, m\}$ enforces monotonicity of the adjusted $p$-values and ensures that the procedure is indeed step-down, that is, one can only reject a particular hypothesis provided all hypotheses with
##############################################
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-509-\end{equation}
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw:510:where $P_0(m) = \bar{Q}_{0,m}(Z(m))$ and $Z=(Z(m): m=1,\ldots, M) \sim Q_0$.
r-bioc-multtest-2.46.0/inst/otherDocs/MTP.Rnw-511-
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-159-
r-bioc-multtest-2.46.0/vignettes/MTP.tex:160:In many testing problems, the submodels concern {\em parameters}, i.e., functions of the data generating distribution $P$, $\Psi(P) = \psi= (\psi(m):m=1,\ldots,M)$, such as means, differences in means, correlations, and parameters in linear models, generalized linear models, survival models, time-series models, dose-response models, etc. One distinguishes between two types of testing problems: {\em one-sided tests}, where $H_0(m) = \mathrm{I}(\psi(m) \leq \psi_0(m))$, and {\em two-sided tests}, where $H_0(m) = \mathrm{I}(\psi(m) =
r-bioc-multtest-2.46.0/vignettes/MTP.tex-161-\psi_0(m))$.
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-198-\end{equation}
r-bioc-multtest-2.46.0/vignettes/MTP.tex:199:where $n_k$, $\bar{X}_{k,n_k}(m)$, and $\sigma_{k,n_k}^2(m)$ denote, respectively, the sample size, sample means, and sample variances, for patients with test status $k$, $k=0,\, 1$. The null hypotheses are rejected, i.e., the corresponding genes are declared differentially expressed, for large values of the test statistics $T_n(m)$.\\
r-bioc-multtest-2.46.0/vignettes/MTP.tex-200-
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-207-\end{equation}
r-bioc-multtest-2.46.0/vignettes/MTP.tex:208:where ${\cal C}_n(m)={\cal C}(T_n,Q_{0n},\alpha)(m)$, $m=1,\ldots,M$, denote possibly random rejection regions. The long notation ${\cal R}(T_n, Q_{0n},\alpha)$ and ${\cal C}(T_n, Q_{0n},\alpha)(m)$ emphasizes that the MTP depends on:
r-bioc-multtest-2.46.0/vignettes/MTP.tex-209-(i) the {\em data}, $X_1, \ldots, X_n$,
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-375-For a broad class of testing problems, such as the test of single-parameter null hypotheses using $t$-statistics (as in Equation (\ref{anal:mult:e:tstat})), the null distribution $Q_0$ is an $M$--variate Gaussian distribution with mean vector zero and covariance matrix $\Sigma^*(P)$: $Q_0 = Q_0(P) \equiv N(0,\Sigma^*(P))$.
r-bioc-multtest-2.46.0/vignettes/MTP.tex:376:For tests of means, where the parameter of interest is the $M$--dimensional mean vector $\Psi(P) = \psi = E[X]$, the estimator $\psi_n$ is simply the $M$--vector of sample averages and $\Sigma^*(P)$ is the correlation matrix of $X \sim P$, $Cor[X]$. More generally, for an asymptotically linear estimator $\psi_n$, $\Sigma^*(P)$ is the correlation matrix of the vector influence curve (IC).
r-bioc-multtest-2.46.0/vignettes/MTP.tex-377-
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-458-\end{equation}
r-bioc-multtest-2.46.0/vignettes/MTP.tex:459:where $Z^{\circ}(m)$ denotes the $m$th ordered component of $Z = (Z(m): m=1,\ldots,M) \sim Q_0$, so that $Z^{\circ}(1) \geq \ldots \geq Z^{\circ}(M)$.
r-bioc-multtest-2.46.0/vignettes/MTP.tex-460-For FWER control ($k=0$), the procedure reduces to the {\em single-step maxT procedure}, based on the {\em maximum test statistic}, $Z^{\circ}(1)$.\\
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-463-{\bf Single-step common-quantile procedure.} The set of rejected hypotheses for the {\em $\theta$--controlling single-step common-quantile procedure} is of the form
r-bioc-multtest-2.46.0/vignettes/MTP.tex:464:${\cal R}_n \equiv \{m: T_n(m)> c_0(m) \}$, where $c_0(m) = Q_{0,m}^{-1}(\delta_0)$ is the $\delta_0$--quantile of the marginal null distribution $Q_{0,m}$ of the $m$th test statistic, i.e., the smallest value $c$ such that $Q_{0,m}(c) = Pr_{Q_0}(Z(m) \leq c) \geq \delta_0$ for $Z \sim Q_0$. Here, $\delta_0$ is chosen as the {\em smallest} (i.e., least conservative) value for which $\theta(F_{R_0}) \leq \alpha$.
r-bioc-multtest-2.46.0/vignettes/MTP.tex-465-
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-470-\end{equation}
r-bioc-multtest-2.46.0/vignettes/MTP.tex:471:where $P_0^{\circ}(m)$ denotes the $m$th ordered component of the $M$--vector of unadjusted $p$-values $(P_0(m): m=1,\ldots,M)$, so that $P_0^{\circ}(1) \leq \ldots \leq P_0^{\circ}(M)$.
r-bioc-multtest-2.46.0/vignettes/MTP.tex-472-For FWER control ($k=0$), one recovers the {\em single-step minP procedure}, based on the {\em minimum unadjusted $p$-value}, $P_0^{\circ}(1)$.
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-494-\end{equation}
r-bioc-multtest-2.46.0/vignettes/MTP.tex:495:where $Z=(Z(m): m=1,\ldots, M) \sim Q_0$.
r-bioc-multtest-2.46.0/vignettes/MTP.tex-496-Taking maxima of the probabilities over $h \in \{1, \ldots, m\}$ enforces monotonicity of the adjusted $p$-values and ensures that the procedure is indeed step-down, that is, one can only reject a particular hypothesis provided all hypotheses with
##############################################
r-bioc-multtest-2.46.0/vignettes/MTP.tex-509-\end{equation}
r-bioc-multtest-2.46.0/vignettes/MTP.tex:510:where $P_0(m) = \bar{Q}_{0,m}(Z(m))$ and $Z=(Z(m): m=1,\ldots, M) \sim Q_0$.
r-bioc-multtest-2.46.0/vignettes/MTP.tex-511-