Fix rendering issues

Author: Susan Vanderplas

_freeze/part-advanced-topics/01-simulation/execute-results/html.json

@@ -246,7 +246,7 @@ plt.show()
 
 Simulation from different distributions can be used to determine which estimators are most appropriate for a given scenario, to determine how likely it is to observe a specific value in a sample of size $n$, and for many other applications. 
 
-::: callout-caution
+:::: callout-caution
 
 ### Example: Mean vs. Median
 
@@ -261,7 +261,7 @@ We know that the MSE is related to the bias and the variance of $\hat\theta$: $\
 
 How can you use this information to determine which estimator is preferable in each $\nu, n$ situation?
 
-::: tabset-panel
+::: panel-tabset
 
 #### R
 
@@ -388,7 +388,7 @@ import matplotlib.pyplot as plt
 Considering the results we obtained, it seems clear that when we have a very low ($\nu<=5$) degrees of freedom, the median is a preferable estimator because it has lower variance; when we have a higher number of degrees of freedom, the mean is a preferable estimator on the basis of variance. 
 Both estimators are asymptotically unbiased, and are unbiased even for small sample sizes when the distribution is symmetric. 
 
-:::
+::::
 
 ## Simulation to test model assumptions