Merge pull request #62 from r-causal/qa-clean-up

Author: malcolmbarrett

exercises/01-whole-game-exercises.qmd

@@ -90,6 +90,7 @@ smk_wt_dag <- dagify(
   # specify causal question:
   exposure = "qsmk", 
   outcome = "wt82_71",
+  coords = time_ordered_coords(),
   # set up labels:
   # here, I'll use the same variable names as the data set, but I'll label them
   # with clearer names
@@ -113,7 +114,8 @@ smk_wt_dag <- dagify(
     "smokeintensity" = "smoking\nintensity",
     "smokeyrs" = "yrs of\nsmoking"
   )
-) |>tidy_dagitty(layout = "star")
+) |>
+  tidy_dagitty()
 
 smk_wt_dag
 ```
@@ -313,9 +315,13 @@ fit_ipw <- function(split, ...) {
   # calculate inverse probability weights
   .df <- propensity_model |>
     augment(type.predict = "response", data = .df) |>
-    mutate(wts = wt_ate(.fitted, qsmk))
+    mutate(wts = wt_ate(
+      .fitted,
+      qsmk, 
+      exposure_type = "binary"
+    ))
   
-  # fit correctly bootsrapped ipw model
+  # fit correctly bootstrapped ipw model
   lm(wt82_71 ~ qsmk, data = .df, weights = wts) |>
     tidy()
 }

exercises/03-ci-with-group-by-and-summarise-exercises.qmd

@@ -77,7 +77,6 @@ sim2 |>
               names_prefix = "x_") |>
   summarise(estimate = x_1 - x_0, .groups = "drop") |>
   summarise(estimate = mean(estimate))  
-
 ```
 
 ## Your Turn 3
@@ -108,8 +107,8 @@ sim3 |>
     values_from = avg_y,  
     names_prefix = "x_"
   ) |>
-  summarise(estimate = x_1 - x_0) 
-
+  summarise(estimate = x_1 - x_0)  |>
+  summarise(estimate = mean(estimate))  
 ```
 
 # Take aways

exercises/04-dags-exercises.qmd

@@ -103,7 +103,7 @@ coffee_cancer_dag_to <- dagify(
   )
 )
 
-_____(_____, use_labels = TRUE, use_text = FALSE)
+_____(_____, use_labels = "label", text = FALSE)
 ```
 
 # Take aways

exercises/05-quartets-exercises.qmd

@@ -43,3 +43,7 @@ For each of the following 4 datasets, fit a linear linear model examining the re
 * `causal_mediator_time`
 * `causal_m_bias_time`
 
+```{r}
+
+```
+

exercises/07-pscores-using-exercises.qmd

@@ -17,7 +17,6 @@ We are interested in examining the relationship between whether there were "Extr
 Below is the propensity score model you created in the previous exercise.
 
 ```{r}
-#| eval: false
 seven_dwarfs <- seven_dwarfs_train_2018 |>
   filter(hour == 9)
 

exercises/08-pscores-diagnostics-exercises.qmd

@@ -18,7 +18,6 @@ We are interested in examining the relationship between whether there were "Extr
 Below is the propensity score model and weights you created in the previous exercise.
 
 ```{r}
-#| eval: false
 seven_dwarfs <- seven_dwarfs_train_2018 |>
   filter(hour == 9)
 

exercises/09-outcome-model-exercises.qmd

@@ -10,6 +10,7 @@ library(tidyverse)
 library(broom)
 library(touringplans)
 library(rsample)
+library(propensity)
 
 seven_dwarfs <- seven_dwarfs_train_2018 |>
   filter(hour == 9)
@@ -56,9 +57,9 @@ ipw_results |>
   mutate(
     estimate = map_dbl(
       boot_fits,
-      # pull the `estimate` for `qsmk` for each fit
+      # pull the `estimate` for `extra_magic_morning` for each fit
       \(.fit) .fit |>
-        filter(term == "qsmk") |>
+        filter(term == "extra_magic_morning") |>
         pull(estimate)
     )
   ) |>
@@ -73,6 +74,8 @@ Calculate the confidence interval
 #| eval: false
 boot_estimate <- ____(____, ____) |>
   filter(term == ____)
+
+boot_estimate
 ```
 
 

exercises/10-continuous-g-computation-exercises.qmd

@@ -184,7 +184,7 @@ fit_gcomp <- function(split, ...) {
   
   
   # calculate ATE
-  bind_cols(predicted_yes, predicted_no) |>
+  bind_cols(predicted_thirty, predicted_sixty) |>
     summarize(
       mean_thirty = mean(thirty_posted_minutes),
       mean_sixty = mean(sixty_posted_minutes),

exercises/11-tipr-exercises.qmd

@@ -12,7 +12,7 @@ We are interested in examining the relationship between whether there were "Extr
 
 ## Your turn
 
-Use the `tip_coef()` function to conduct a sensitivity analysis for the estimate from your previous exercises.
+Use the `tip_coef()` function to conduct a sensitivity analysis for the estimate from your previous exercises. Use the lower bound of the confidence interval for the effect and `0.1` for the exposure-confounder effect.
 
 ```{r}
 

exercises/13-bonus-selection-bias-exercises.qmd

@@ -9,6 +9,7 @@ library(tidyverse)
 library(broom)
 library(propensity)
 library(causaldata)
+library(rsample)
 ```
 
 In this example, we'll consider loss to follow-up in the NHEFS study. We'll use the binary exposure we used earlier in the workshop: does quitting smoking (`smk`) increase weight (`wt82_71`)? This time, however, we'll adjust for loss to followup (people who dropped out of the study between observation periods) using inverse probability of censoring weights.