run typos

Author: getzze

README.md

@@ -61,7 +61,7 @@ julia>] add RobustModels#main;
 
 ## Usage
 
-The prefered way of performing robust regression is by calling the `rlm` function:
+The preferred way of performing robust regression is by calling the `rlm` function:
 
 `m = rlm(X, y, MEstimator{TukeyLoss}(); initial_scale=:mad)`
 
@@ -156,7 +156,7 @@ Several loss functions are implemented:
 - `CauchyLoss`: `ρ(r) = log(1+(r/c)²)`, non-convex estimator, that also corresponds to a Student's-t distribution (with fixed degree of freedom). It suppresses outliers more strongly but it is not sure to converge.
 - `GemanLoss`: `ρ(r) = ½ (r/c)²/(1 + (r/c)²)`, non-convex and bounded estimator, it suppresses outliers more strongly.
 - `WelschLoss`: `ρ(r) = ½ (1 - exp(-(r/c)²))`, non-convex and bounded estimator, it suppresses outliers more strongly.
-- `TukeyLoss`: `ρ(r) = if r<c; ⅙(1 - (1-(r/c)²)³) else ⅙ end`, non-convex and bounded estimator, it suppresses outliers more strongly and it is the prefered estimator for most cases.
+- `TukeyLoss`: `ρ(r) = if r<c; ⅙(1 - (1-(r/c)²)³) else ⅙ end`, non-convex and bounded estimator, it suppresses outliers more strongly and it is the preferred estimator for most cases.
 - `YohaiZamarLoss`: `ρ(r)` is quadratic for `r/c < 2/3` and is bounded to 1; non-convex estimator, it is optimized to have the lowest bias for a given efficiency.
 
 The value of the tuning constants `c` are optimized for each estimator so the M-estimators have a high efficiency of 0.95. However, these estimators have a low breakdown point.
@@ -184,9 +184,9 @@ the two loss functions should be the same but with different tuning constants.
 ### MQuantile-estimators
 
 Using an asymmetric variant of the `L1Estimator`, quantile regression is performed
-(although the `QuantileRegression` solver should be prefered because it gives an exact solution).
-Identically, with an M-estimator using an asymetric version of the loss function,
-a generalization of quantiles is obtained. For instance, using an asymetric `L2Loss` results in _Expectile Regression_.
+(although the `QuantileRegression` solver should be preferred because it gives an exact solution).
+Identically, with an M-estimator using an asymmetric version of the loss function,
+a generalization of quantiles is obtained. For instance, using an asymmetric `L2Loss` results in _Expectile Regression_.
 
 ### Robust Ridge regression
 
@@ -215,7 +215,7 @@ This package derives from the [RobustLeastSquares](https://github.com/FugroRoame
 package for the initial implementation, especially for the Conjugate Gradient
 solver and the definition of the M-Estimator functions.
 
-Credits to the developpers of the [GLM](https://github.com/JuliaStats/GLM.jl)
+Credits to the developers of the [GLM](https://github.com/JuliaStats/GLM.jl)
 and [MixedModels](https://github.com/JuliaStats/MixedModels.jl) packages
 for implementing the Iteratively Reweighted Least Square algorithm.
 

docs/src/manual.md

@@ -32,7 +32,7 @@ Supported loss functions are:
 - [`CauchyLoss`](@ref): `ρ(r) = log(1+(r/c)²)`, non-convex estimator, that also corresponds to a Student's-t distribution (with fixed degree of freedom). It suppresses outliers more strongly but it is not sure to converge.
 - [`GemanLoss`](@ref): `ρ(r) = ½ (r/c)²/(1 + (r/c)²)`, non-convex and bounded estimator, it suppresses outliers more strongly.
 - [`WelschLoss`](@ref): `ρ(r) = ½ (1 - exp(-(r/c)²))`, non-convex and bounded estimator, it suppresses outliers more strongly.
-- [`TukeyLoss`](@ref): `ρ(r) = if r<c; ⅙(1 - (1-(r/c)²)³) else ⅙ end`, non-convex and bounded estimator, it suppresses outliers more strongly and it is the prefered estimator for most cases.
+- [`TukeyLoss`](@ref): `ρ(r) = if r<c; ⅙(1 - (1-(r/c)²)³) else ⅙ end`, non-convex and bounded estimator, it suppresses outliers more strongly and it is the preferred estimator for most cases.
 - [`YohaiZamarLoss`](@ref): `ρ(r)` is quadratic for `r/c < 2/3` and is bounded to 1; non-convex estimator, it is optimized to have the lowest bias for a given efficiency.
 
 An estimator is constructed from an estimator type and a loss, e.g. `MEstimator{TukeyLoss}()`.

src/estimators.jl

@@ -224,7 +224,7 @@ function scale_estimate(est::E, res; kwargs...) where {E<:MEstimator}
     return scale_estimate(est.loss, res; kwargs...)
 end
 
-"`L1Estimator` is a shorthand name for `MEstimator{L1Loss}`. Using exact QuantileRegression should be prefered."
+"`L1Estimator` is a shorthand name for `MEstimator{L1Loss}`. Using exact QuantileRegression should be preferred."
 const L1Estimator = MEstimator{L1Loss}
 
 "`L2Estimator` is a shorthand name for `MEstimator{L2Loss}`, the non-robust OLS."

src/losses.jl

@@ -439,7 +439,7 @@ estimator_high_efficiency_constant(::Type{CauchyLoss}) = 2.385
 estimator_high_breakdown_point_constant(::Type{CauchyLoss}) = 1.468
 
 """
-The non-convex Geman-McClure for strong supression of outliers and does not guarantee a unique solution.
+The non-convex Geman-McClure for strong suppression of outliers and does not guarantee a unique solution.
 For the S-Estimator, it is equivalent to the Cauchy loss.
 ψ(r) = r / (1 + r^2)^2
 """
@@ -469,7 +469,7 @@ estimator_high_breakdown_point_constant(::Type{GemanLoss}) = 0.61200
 
 
 """
-The non-convex Welsch for strong supression of outliers and does not guarantee a unique solution
+The non-convex Welsch for strong suppression of outliers and does not guarantee a unique solution
 ψ(r) = r * exp(-r^2)
 """
 struct WelschLoss <: BoundedLossFunction