typo fixes

Author: ngreifer

NEWS.md

@@ -11,7 +11,7 @@ WeightIt News and Updates
 
 * With `method = "cbps"`, `estimand` can now be set to `"ATO"` for binary and multi-category treatments. For binary treatments with the default link, this will yield identical weights to using `method = "glm"` with `estimand = "ATO"`.
 
-* Two new links can be supplied with `method = "glm", `"cbps"`, and `"ipt"`: `"loglog"` for the log-log link and `"clog"` for the complementary log link. The `link` argument can also now be supplied as a `link-glm` object (e.g., the output of a call to `make.link()`). This allows for more flexibility in the link function used to estimate the propensity score.
+* Two new links can be supplied with `method = "glm"`, `"cbps"`, and `"ipt"`: `"loglog"` for the log-log link and `"clog"` for the complementary log link. The `link` argument can also now be supplied as a `link-glm` object (e.g., the output of a call to `make.link()`). This allows for more flexibility in the link function used to estimate the propensity score.
 
 * A new `solver` argument can be supplied to `weightit()` with `method = "ebal"` and with `method = "cbps"` with `over = FALSE` (the default); this argument controls whether to use `rootSolve::multiroot()` or `stats::optim()` solve the optimization problem for the weights. `multiroot()` is used by default when `rootSolve` is installed as it is quicker and more accurate.
 
@@ -161,7 +161,7 @@ WeightIt News and Updates
 
 * Added a new function, `glm_weightit()` (along with wrapper `lm_weightit()`) and associated methods for fitting generalized linear models in the weighted sample, with the option of accounting for estimation of the weights in computing standard errors via M-estimation or two forms of bootstrapping. `glm_weightit()` also supports multinomial logistic regression in addition to all models supported by `glm()`. Cluster-robust standard errors are supported, and output is compatible with any functions that accept `glm()` objects. Not all weighting methods support M-estimation, but for those that do, a new component is added to the `weightit` output object. Currently, GLM propensity scores, entropy balancing, just-identified CBPS, and inverse probability tilting (described below) support M-estimation-based standard errors with `glm_weightit()`.
 
-* Added inverse probability tilting (IPT) as described by Graham, Pinto, and Egel (2012), which can be requested by setting `method = "ipt"`. Thus is similar to entropy balancing and CBPS in that it enforces exact balance and yields a propensity score, but has some theoretical advantages to both methods. IPT does not rely on any other packages and runs very quickly.
+* Added inverse probability tilting (IPT) as described by Graham, Pinto, and Egel (2012), which can be requested by setting `method = "ipt"`. This is similar to entropy balancing and CBPS in that it enforces exact balance and yields a propensity score, but has some theoretical advantages to both methods. IPT does not rely on any other packages and runs very quickly.
 
 * Estimating covariate balancing propensity score weights (i.e., `method = "cbps"`) no longer depends on the `CBPS` package. The default is now the just-identified versions of the method; the over-identified version can be requested by setting `over = TRUE`. The ATT for multi-category treatments is now supported, as are arbitrary numbers of treatment groups (`CBPS` only natively support up to 4 groups and only the ATE for multi-category treatments). For binary treatments, generalized linear models other than logistic regression are now supported (e.g., probit or Poisson regression).