more math mode fixes

06-sl3.Rmd

@@ -15,17 +15,16 @@ Coyle, Nima Hejazi, Ivana Malenica, Rachael Phillips, and Oleg Sofrygin_.
 
 <!--
 IM:
-Since we introduced the concept of loss & risk in the previous chapter, 
-it should be ok use it instead of "performance metric". I might be a 
-bit confusing to change notation many times. 
-Define base learner
-Shouldn't objective 4 be before talking about screeners ect?
+Since we introduced the concept of loss & risk in the previous chapter, it
+should be ok use it instead of "performance metric". I might be a bit confusing
+to change notation many times. Define base learner Shouldn't objective 4 be
+before talking about screeners ect?
 -->
 
 By the end of this chapter you will be able to:
 
 1. Select a performance metric that is optimized by the true prediction
-   function, or define the true prediction prediction of interest as the 
+   function, or define the true prediction prediction of interest as the
    optimizer of the performance metric.
 2. Assemble a diverse set ("library") of learners to be considered in the super
    learner. In particular, you should be able to:
@@ -1853,7 +1852,7 @@ location-scale_ family, that is, in which $p_n(Y \mid X) = \rho((Y - \mu_n(X)) /
 potential disadvantages (e.g., the restriction on the density's functional form
 could lead to misspecification bias), this strategy is flexible in that it
 allows for arbitrary machine learning algorithms to be used in estimating the
-conditional mean of $Y$ given $X$, \mu(X) = \E(Y \mid X)$, and the conditional
+conditional mean of $Y$ given $X$, $\mu(X) = \E(Y \mid X)$, and the conditional
 variance of $Y$ given $X$, $\sigma(X) = \E[(Y - \mu(X))^2 \mid X]$.
 
 In settings with limited data, the additional structure imposed by the