Update SAS Code Execution and Rendering

Comp/r-east_gsd_tte.qmd

@@ -2,8 +2,6 @@
 title: "R vs EAST vs SAS: Group sequential design"
 editor_options: 
   chunk_output_type: console
-execute: 
-  eval: false
 ---
 
 ## Introduction
@@ -30,7 +28,6 @@ We assume that a GSD is utilized for progression-free survival (PFS) endpoint. I
 Further design assumptions are as follows:
 
 ```{r}
-#| eval: true
 # PFS HR=0.6
 hr1_pfs <- 0.6
 # median PFS of 9.4 months in the control arm
@@ -62,7 +59,6 @@ Note that, in EAST the number of target events is reported as an integer, howeve
 For ease of comparison the results from EAST are summarized below:
 
 ```{r}
-#| eval: true
 #| echo: false
 #| warning: false
 library(flextable)
@@ -106,7 +102,6 @@ pfs_east |>
 -   gsDesign code to reproduce the above EAST results:
 
 ```{r}
-#| eval: true
 #| warning: false
 library(gsDesign)
 
@@ -136,7 +131,6 @@ pfs_gsDesign |>
 -   gsDesign vs EAST comparison using absolute differences:
 
 ```{r}
-#| eval: true
 #| echo: false
 digit_comp <- 4
 pfs_gsDesign |>
@@ -191,7 +185,6 @@ pfs_gsDesign |>
 -   Note that, here `gsDesign2::gs_power_ahr()` is used given the number of target events for each analysis based on EAST results.
 
 ```{r}
-#| eval: true
 #| echo: false
 #helper function to align the gsDesign2 summary with gsDesign summary
 as_gs <- function(xnph) {
@@ -318,7 +311,6 @@ as_gs <- function(xnph) {
 ```
 
 ```{r}
-#| eval: true
 #| warning: false
 #| message: false
 library(gsDesign2)
@@ -365,7 +357,6 @@ pfs_gsDesign2 |>
 -   gsDesign2 vs EAST comparison using absolute differences:
 
 ```{r}
-#| eval: true
 #| echo: false
 pfs_gsDesign2 |>
   as_gs() |>
@@ -418,7 +409,6 @@ pfs_gsDesign2 |>
 -   rpact code to reproduce the above EAST results appears below.
 
 ```{r}
-#| eval: true
 #| warning: false
 library(rpact)
 
@@ -451,7 +441,6 @@ kable(summary(pfs_rpact))
 
 ```{r}
 #| echo: false
-#| eval: true
 
 pcross_h1_eff <- cumsum(pfs_rpact$rejectPerStage)
 pcross_h1_fut <- pfs_rpact$futilityPerStage[1]
@@ -548,7 +537,7 @@ pfs_rpact_sum |>
 
 -   SAS code to reproduce the above rpact results appears below.
 
-```{sas}
+```sas
 PROC SEQDESIGN BOUNDARYSCALE=MLE ERRSPEND;
    DESIGN NSTAGES=2 
           INFO=CUM(0.748936170212766 1.0) 
@@ -572,7 +561,6 @@ RUN;
 The following shows the events (D) and required sample sizes (N) for IA and FA.
 
 ```{r}
-#| eval: true
 #| echo: false
 #| fig-align: center
 #| out-width: 100%
@@ -585,7 +573,7 @@ Please note that the `BOUNDARYSCALE=MLE | SCORE | STDZ | PVALUE` options display
 
 SAS doesn't provide a boundary information with HR, so the HR boundaries is obtained from the MLE boundaries (as MLE $=\hat{\theta}=-log(\text{HR})$, see [SAS User's Guide: Test for Two Survival Distributions with a Log-Rank Test](https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.5/statug/statug_seqdesign_details52.htm#statug.seqdesign.cseqdlogrank)) via the following code.
 
-```{sas}
+```sas
 DATA BHR;
    SET BMLE;
    Bound_UA_HR=exp(-Bound_UA);
@@ -599,7 +587,6 @@ RUN;
 The HR boundaries are shown below.
 
 ```{r}
-#| eval: true
 #| echo: false
 #| fig-align: center
 #| out-width: 100%
@@ -611,7 +598,6 @@ knitr::include_graphics(
 The results calculated by SAS are presneted in the table below. Please note that SAS doesn't report the probablities $P(Cross | HR=1)$ and $P(Cross | HR=0.6)$, resulting in empty cells for these results in the table.
 
 ```{r}
-#| eval: true
 #| echo: false
 #| warning: false
 library(flextable)
@@ -648,7 +634,6 @@ pfs_sas |>
 -   SAS vs rapct comparison using absolute differences:
 
 ```{r}
-#| eval: true
 #| echo: false
 #| warning: false
 pfs_rpact_sum |>
@@ -694,7 +679,6 @@ pfs_rpact_sum |>
 -   SAS vs EAST comparison using absolute differences:
 
 ```{r}
-#| eval: true
 #| echo: false
 pfs_sas_sum <- tibble::tibble(
   sas_eff = pfs_sas$eff_125,

Comp/r-sas-python_survey-stats-summary.qmd

@@ -1,8 +1,6 @@
 ---
 title: "R vs SAS vs Python Survey Summary Statistics"
 bibliography: survey-stats-summary.bib
-execute: 
-  eval: false
 ---
 
 This document will compare the survey summary statistics functionality in SAS (available through SAS/STAT), R (available from the [`{survey}`](%5B%60%7Bsurvey%7D%60%5D(https://r-survey.r-forge.r-project.org/survey/html/api.html)) package), and Python (available from the [`samplics`](https://samplics-org.github.io/samplics/) package), highlighting differences in methods and results. Only the default Taylor series linearisation method for calculating variances is used in all languages. A more detailed comparison between R and SAS for specific methods and use-cases is available in [@2017_YRBS], [@so2020modelling], or [@adamico_2009]. For a general guide to survey statistics, which has companion guides for both R and SAS, see [@Lohr_2022].
@@ -31,7 +29,6 @@ For the full R, SAS, and Python code and results used for this comparison, see b
 ## R
 
 ```{r}
-#| eval: true
 #| message: false
 #| warning: false
 library(survey)
@@ -125,7 +122,7 @@ print(list(
 
 ## SAS
 
-```{sas}
+```sas
 * Mean, sum quantile of HI_CHOL;
 proc surveymeans data=nhanes mean sum clm quantile=(0.025 0.5 0.975);
     cluster SDMVPSU;
@@ -277,7 +274,6 @@ run;
 ## Python
 
 ```{python}
-#| eval: true
 import pandas as pd
 from samplics import TaylorEstimator
 from samplics.utils.types import PopParam
@@ -363,7 +359,6 @@ print(
 `samplics` in Python does not have a method for calculating quantiles, and in R and SAS the available methods lead to different results. To demonstrate the differences in calculating quantiles, we will use the `apisrs` dataset from the `survey` package in R [@API_2000].
 
 ```{r}
-#| eval: true
 #| message: false
 library(survey)
 
@@ -377,7 +372,7 @@ In SAS, PROC SURVEYMEANS will calculate quantiles of specific probabilities as y
 
 The method and results from SAS are as follows:
 
-```{sas}
+```sas
 proc surveymeans data=apisrs total=6194 quantile=(0.025 0.5 0.975);
     var growth;
 run;
@@ -407,7 +402,6 @@ run;
 If in R we use the default `qrule="math"` (equivalent to `qrule="hf1"` and matches `type=1` in the `quantile` function for unweighted data) along with the default `interval.type="mean"`, we get the following results:
 
 ```{r}
-#| eval: true
 srs_design <- survey::svydesign(data = apisrs, id = ~1, fpc = ~fpc, )
 
 survey::svyquantile(
@@ -422,7 +416,6 @@ survey::svyquantile(
 Here we can see that the quantiles, confidence intervals, and standard errors do not match SAS. From testing, none of the available `qrule` methods match SAS for the quantile values, so it is recommended to use the default values unless you have need of some of the other properties of different quantile definitions - see [`vignette("qrule", package="survey")`](https://cran.r-project.org/web/packages/survey/vignettes/qrule.pdf) for more detail. If an exact match to SAS is required, then the `svyquantile` function allows for passing a custom function to the `qrule` argument to define your own method for calculating quantiles. Below is an example that will match SAS:
 
 ```{r}
-#| eval: true
 sas_qrule <- function(x, w, p) {
   # Custom qrule to match SAS, based on survey::oldsvyquantile's internal method
   if (any(is.na(x))) {
@@ -459,7 +452,6 @@ sas_quants
 Note that although the quantiles and standard errors match, the confidence intervals still do not match SAS. For this another custom calculation is required, based on the formula used in SAS:
 
 ```{r}
-#| eval: true
 sas_quantile_confint <- function(newsvyquantile, level = 0.05, df = Inf) {
   q <- coef(newsvyquantile)
   se <- survey::SE(newsvyquantile)
@@ -503,7 +495,6 @@ In contrast, the `samplics` package in Python is still early in development, and
 
 ::: {.callout-note collapse="true" title="Session Info"}
 ```{r}
-#| eval: true
 #| echo: false
 si <- sessioninfo::session_info("survey", dependencies = FALSE)
 # If reticulate is used, si will include python info. However, this doesn't

Comp/r-sas-summary-stats.qmd

@@ -1,6 +1,5 @@
 ---
 title: "Deriving Quantiles or Percentiles in R vs SAS"
-eval: false
 ---
 
 ### Data
@@ -15,7 +14,7 @@ c(10, 20, 30, 40, 150, 160, 170, 180, 190, 200)
 
 Assuming the data above is stored in the variable `aval` within the dataset `adlb`, the 25th and 40th percentiles could be calculated using the following code.
 
-```{sas}
+```sas
 proc univariate data=adlb;
   var aval;
   output out=stats pctlpts=25 40 pctlpre=p;
@@ -38,14 +37,14 @@ The procedure has the option `PCTLDEF` which allows for five different percentil
 The 25th and 40th percentiles of `aval` can be calculated using the `quantile` function.
 
 ```{r}
+#| eval: false
 quantile(adlb$aval, probs = c(0.25, 0.4))
 ```
 
 This gives the following output.
 
 ```{r}
 #| echo: false
-#| eval: true
 adlb <- data.frame(aval = c(10, 20, 30, 40, 150, 160, 170, 180, 190, 200))
 quantile(adlb$aval, probs = c(0.25, 0.4))
 ```
@@ -59,14 +58,14 @@ The default percentile definition used by the UNIVARIATE procedure in SAS finds
 It is possible to get the quantile function in R to use the same definition as the default used in SAS, by specifying `type=2`.
 
 ```{r}
+#| eval: false
 alquantile(adlb$aval, probs = c(0.25, 0.4), type = 2)
 ```
 
 This gives the following output.
 
 ```{r}
 #| echo: false
-#| eval: true
 quantile(adlb$aval, probs = c(0.25, 0.4), type = 2)
 ```
 

Comp/r-sas_anova.qmd

@@ -1,7 +1,5 @@
 ---
 title: "R vs SAS Linear Models"
-execute: 
-  eval: false
 ---
 
 # R vs. SAS ANOVA
@@ -31,7 +29,6 @@ The following table provides an overview of the support and results comparabilit
 In order to get the ANOVA model fit and sum of squares you can use the `anova` function in the `stats` package.
 
 ```{r}
-#| eval: true
 library(emmeans)
 drug_trial <- read.csv("../data/drug_trial.csv")
 
@@ -43,7 +40,6 @@ lm_model |>
 It is recommended to use the `emmeans` package to get the contrasts between R.
 
 ```{r}
-#| eval: true
 lm_model |>
   emmeans("drug") |>
   contrast(
@@ -56,7 +52,7 @@ lm_model |>
 
 In SAS, all contrasts must be manually defined, but the syntax is largely similar in both.
 
-```{sas}
+```sas
 proc glm data=work.mycsv;
    class drug;
    model post = pre drug / solution;
@@ -66,7 +62,6 @@ run;
 ```
 
 ```{r}
-#| eval: true
 #| echo: false
 #| fig-align: center
 #| out-width: 75%
@@ -119,7 +114,7 @@ Provided below is a detailed comparison of the results obtained from both SAS an
 
 Note, however, that there are some cases where the scale of the parameter estimates between SAS and R is off, though the test statistics and p-values are identical. In these cases, we can adjust the SAS code to include a divisor. As far as we can tell, this difference only occurs when using the predefined Base R contrast methods like `contr.helmert`.
 
-```{sas}
+```sas
 proc glm data=work.mycsv;
    class drug;
    model post = pre drug / solution;

Comp/r-sas_chi-sq.qmd

@@ -1,7 +1,5 @@
 ---
 title: "R/SAS Chi-Squared and Fisher's Exact Comparision"
-execute: 
-  eval: false
 ---
 
 # Chi-Squared Test
@@ -26,7 +24,6 @@ For an r x c table (where r is the number of rows and c the number of columns),
 For this example we will use data about cough symptoms and history of bronchitis.
 
 ```{r}
-#| eval: true
 bronch <- matrix(c(26, 247, 44, 1002), ncol = 2)
 row.names(bronch) <- c("cough", "no cough")
 colnames(bronch) <- c("bronchitis", "no bronchitis")
@@ -36,13 +33,12 @@ bronch
 To a chi-squared test in R you will use the following code.
 
 ```{r}
-#| eval: true
 stats::chisq.test(bronch)
 ```
 
 To run a chi-squared test in SAS you used the following code.
 
-```{sas}
+```sas
 proc freq data=proj1.bronchitis;
 tables Cough*Bronchitis / chisq;
 run;
@@ -51,7 +47,6 @@ run;
 The result in the "Chi-Square" section of the results table in SAS will not match R, in this case it is 12.1804 with a p-value of 0.0005. This is because by default R does a Yates continuity adjustment for 2x2 tables. To change this set `correct` to false.
 
 ```{r}
-#| eval: true
 stats::chisq.test(bronch, correct = FALSE)
 ```