rs_stats::regression::multiple_linear_regression

Struct MultipleLinearRegression

pub struct MultipleLinearRegression<T = f64>where
    T: Float + Debug + Default + Serialize,{
    pub coefficients: Vec<T>,
    pub r_squared: T,
    pub adjusted_r_squared: T,
    pub standard_error: T,
    pub n: usize,
    pub p: usize,
}

Expand description

Multiple linear regression model that fits a hyperplane to multivariate data points.

Fields§

§coefficients: Vec<T>

Regression coefficients, including intercept as the first element

§r_squared: T

Coefficient of determination (R²) - goodness of fit

§adjusted_r_squared: T

Adjusted R² which accounts for the number of predictors

§standard_error: T

Standard error of the estimate

§n: usize

Number of data points used for regression

§p: usize

Number of predictor variables (excluding intercept)

Implementations§

Source §

impl<T> MultipleLinearRegression<T>
where T: Float + Debug + Default + NumCast + Serialize + for<'de> Deserialize<'de>,

Source

pub fn new() -> Self

Create a new multiple linear regression model without fitting any data

Source

pub fn fit<U, V>( &mut self, x_values: &[Vec], y_values: &[V], ) -> StatsResult<()>
where U: NumCast + Copy, V: NumCast + Copy,

Fit a multiple linear regression model to the provided data

§Arguments

x_values - 2D array where each row is an observation and each column is a predictor
y_values - Dependent variable values (observations)

§Returns

StatsResult<()> - Ok if successful, Err with StatsError if the inputs are invalid

§Errors

Returns StatsError::EmptyData if input arrays are empty. Returns StatsError::DimensionMismatch if X and Y arrays have different lengths. Returns StatsError::InvalidInput if rows in X have inconsistent lengths. Returns StatsError::ConversionError if value conversion fails. Returns StatsError::MathematicalError if the linear system cannot be solved.

Source

pub fn predict(&self, x: &[U]) -> StatsResult<T>
where U: NumCast + Copy,

Predict y value for a given set of x values using the fitted model

§Arguments

x - Vector of x values for prediction (must match the number of features used during fitting)

§Returns

StatsResult<T> - The predicted y value

§Errors

Returns StatsError::NotFitted if the model has not been fitted (coefficients is empty). Returns StatsError::DimensionMismatch if the number of features doesn’t match the model (x.len() != p). Returns StatsError::ConversionError if type conversion fails.

§Examples

use rs_stats::regression::multiple_linear_regression::MultipleLinearRegression;

let mut model = MultipleLinearRegression::<f64>::new();
let x = vec![
    vec![1.0, 2.0],
    vec![2.0, 1.0],
    vec![3.0, 3.0],
    vec![4.0, 2.0],
];
let y = vec![5.0, 4.0, 9.0, 8.0];
model.fit(&x, &y).unwrap();

let prediction = model.predict(&[3.0, 4.0]).unwrap();

Source

pub fn predict_many(&self, x_values: &[Vec]) -> StatsResult<Vec<T>>
where U: NumCast + Copy,

Calculate predictions for multiple observations

§Arguments

x_values - 2D array of feature values for prediction

§Returns

StatsResult<Vec<T>> - Vector of predicted y values

§Errors

Returns StatsError::NotFitted if the model has not been fitted. Returns an error if any prediction fails (dimension mismatch or conversion error).

§Examples

use rs_stats::regression::multiple_linear_regression::MultipleLinearRegression;

let mut model = MultipleLinearRegression::<f64>::new();
let x = vec![
    vec![1.0, 2.0],
    vec![2.0, 1.0],
    vec![3.0, 3.0],
    vec![4.0, 2.0],
];
let y = vec![5.0, 4.0, 9.0, 8.0];
model.fit(&x, &y).unwrap();

let predictions = model.predict_many(&[vec![3.0, 4.0], vec![5.0, 6.0]]).unwrap();
assert_eq!(predictions.len(), 2);