Struct StatisticalComputingLibrary 

pub struct StatisticalComputingLibrary { /* private fields */ }

Statistical Computing Library Manager

Implementations

impl StatisticalComputingLibrary

pub fn new() -> Self

Create new statistical computing library

pub fn initialize(&mut self) -> Result<(), StatisticalError>

Initialize the library

pub fn create_dataset( &mut self, dataset_id: String, data: Vec<Vec<DataValue>>, column_names: Vec<String>, column_types: Vec<DataType>, privacy_level: PrivacyLevel, ) -> Result<Dataset, StatisticalError>

Create a new dataset

pub fn mean( &mut self, dataset_id: &str, column: &str, privacy_preserved: bool, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Compute mean of a column

pub fn median( &mut self, dataset_id: &str, column: &str, privacy_preserved: bool, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Compute median of a column

pub fn variance( &mut self, dataset_id: &str, column: &str, sample: bool, privacy_preserved: bool, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Compute variance of a column

pub fn correlation( &mut self, dataset_id: &str, column1: &str, column2: &str, method: CorrelationMethod, privacy_preserved: bool, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Compute correlation between two columns

pub fn t_test( &mut self, dataset_id: &str, column: &str, hypothesis_type: HypothesisType, privacy_preserved: bool, ) -> Result<StatisticalAnalysisResult<TTestResult>, StatisticalError>

Perform t-test

pub fn histogram( &mut self, dataset_id: &str, column: &str, bins: usize, privacy_preserved: bool, ) -> Result<StatisticalAnalysisResult<HistogramResult>, StatisticalError>

Generate histogram

pub fn standard_deviation( &self, dataset_id: &str, column: &str, sample: bool, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Standard deviation (sample = true → Bessel-corrected).

pub fn skewness( &self, dataset_id: &str, column: &str, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Sample skewness (Fisher-Pearson).

pub fn kurtosis( &self, dataset_id: &str, column: &str, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Excess kurtosis.

pub fn mode( &self, dataset_id: &str, column: &str, ) -> Result<ModeResult, StatisticalError>

Modal value + its frequency.

pub fn quantile( &self, dataset_id: &str, column: &str, q: f64, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

The q-quantile (q ∈ [0,1]) via linear interpolation between order statistics. percentile is the same with p ∈ [0,100].

pub fn covariance( &self, dataset_id: &str, column_x: &str, column_y: &str, sample: bool, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Covariance between two columns (sample = true → divide by n-1).

pub fn linear_regression( &self, dataset_id: &str, column_x: &str, column_y: &str, ) -> Result<LinearRegression, StatisticalError>

Ordinary-least-squares simple linear regression y ~ x with full inferential statistics (slope/intercept, R², standard errors, t, p).

pub fn polynomial_regression( &self, dataset_id: &str, column_x: &str, column_y: &str, degree: usize, ) -> Result<PolynomialFit, StatisticalError>

Polynomial regression y ~ poly(x, degree) by least squares (normal equations). Returns ascending-power coefficients and in-sample R².

pub fn logistic_regression( &self, dataset_id: &str, feature_columns: &[&str], label_column: &str, fit_intercept: bool, ) -> Result<GlmModel, StatisticalError>

Binary logistic regression by IRLS (label in {0,1}), with Wald standard errors, z-statistics and p-values on the coefficients.

pub fn anova( &self, dataset_id: &str, group_columns: &[&str], ) -> Result<AnovaResult, StatisticalError>

One-way ANOVA across the named columns (each column is a group). Groups may have unequal lengths; Null cells are dropped per column.

pub fn chi_square_gof( &self, dataset_id: &str, observed_column: &str, expected_column: Option<&str>, ) -> Result<ChiSquareResult, StatisticalError>

Chi-square goodness-of-fit test: observed counts vs. expected counts. If expected_column is None, a uniform expectation is used.

pub fn chi_square_independence( &self, dataset_id: &str, columns: &[&str], ) -> Result<ChiSquareResult, StatisticalError>

Chi-square test of independence over a contingency table whose columns are the named columns (each row of the table is one dataset row).

pub fn autocorrelation( &self, dataset_id: &str, column: &str, lag: usize, ) -> Result<StatisticalAnalysisResult<f64>, StatisticalError>

Autocorrelation of a column at the given lag (biased estimator).

pub fn moving_average( &self, dataset_id: &str, column: &str, window: usize, ) -> Result<Vec<f64>, StatisticalError>

Simple moving average of a column with the given window.

pub fn exponential_smoothing( &self, dataset_id: &str, column: &str, alpha: f64, ) -> Result<Vec<f64>, StatisticalError>

Single exponential smoothing of a column with factor alpha ∈ (0,1].

pub fn kmeans( &self, dataset_id: &str, feature_columns: &[&str], k: usize, max_iter: usize, seed: u64, ) -> Result<KMeansModel, StatisticalError>

K-means clustering over the named feature columns. Returns the fitted model (centroids, per-point labels, inertia, convergence).

pub fn linear_svm( &self, dataset_id: &str, feature_columns: &[&str], label_column: &str, c: f64, ) -> Result<SvmFitResult, StatisticalError>

Soft-margin linear SVM over the named feature columns with a boolean label column (non-zero = positive class). Returns a fit summary with support-vector count and in-sample accuracy.

pub fn random_forest( &self, dataset_id: &str, feature_columns: &[&str], target_column: &str, n_trees: usize, classifier: bool, seed: u64, ) -> Result<RandomForestFitResult, StatisticalError>

Random-forest fit over the named feature columns and a target column. classifier = true fits a classification forest (integer labels) and reports in-sample accuracy; otherwise a regression forest reporting R².

pub fn get_performance_stats(&self) -> SystemMetrics

Get performance statistics

pub fn list_datasets(&self) -> Vec<String>

List all datasets

pub fn get_dataset_info(&self, dataset_id: &str) -> Option<DatasetMetadata>

Get dataset information