feat: add multivariate interface

src/stats/copulas/bivariate.rs

@@ -13,15 +13,13 @@ pub mod gumbel;
 pub mod independence;
 
 #[derive(Debug, Clone, Copy)]
-pub enum CopulaType {
+pub(super) enum CopulaType {
   Clayton,
   Frank,
   Gumbel,
   Independence,
 }
 
-const EPSILON: f64 = 1e-12;
-
 pub trait Bivariate {
   fn r#type(&self) -> CopulaType;
 
@@ -128,9 +126,7 @@ pub trait Bivariate {
   fn pdf(&self, X: &Array2<f64>) -> Result<Array1<f64>, Box<dyn Error>>;
 
   fn log_pdf(&self, X: &Array2<f64>) -> Result<Array1<f64>, Box<dyn Error>> {
-    let pdf = self.pdf(X)?;
-    let log_pdf = pdf.mapv(|val| (val + 1e-32).ln());
-    Ok(log_pdf)
+    Ok(self.pdf(X)?.ln())
   }
 
   fn cdf(&self, X: &Array2<f64>) -> Result<Array1<f64>, Box<dyn Error>>;

src/stats/copulas/multivariate.rs

@@ -1,3 +1,31 @@
+use std::error::Error;
+
+use ndarray::{Array1, Array2};
+
 pub mod gaussian;
 pub mod tree;
 pub mod vine;
+
+pub(super) enum CopulaType {
+  Gaussian,
+  Tree,
+  Vine,
+}
+
+pub trait Multivariate {
+  fn r#type(&self) -> CopulaType;
+
+  fn sample(&self, n: usize) -> Result<Array2<f64>, Box<dyn Error>>;
+
+  fn fit(&mut self, X: Array2<f64>) -> Result<(), Box<dyn Error>>;
+
+  fn check_fit(&self, X: &Array2<f64>) -> Result<(), Box<dyn Error>>;
+
+  fn pdf(&self, X: Array2<f64>) -> Result<Array1<f64>, Box<dyn Error>>;
+
+  fn log_pdf(&self, X: Array2<f64>) -> Result<Array1<f64>, Box<dyn Error>> {
+    Ok(self.pdf(X)?.ln())
+  }
+
+  fn cdf(&self, X: Array2<f64>) -> Result<Array1<f64>, Box<dyn Error>>;
+}