Build a Time Series Crate in Rust and Publish to Cargo

Develop a time series autocorrelation module in Rust by implementing a mean function, computing numerator and denominator with lag, and exposing a public autocorrelation function for testing with cargo.

Explore a simplified stationarity check for time series using rolling mean and rolling standard deviation, via an augmented dickey-fuller approach implemented in Rust.

Explore the math behind linear regression for identifying trend patterns, derive the line y = m x + c, and compute slope, intercept, and means to project future values.

Convert linear regression math into Rust code to detect time series trends. Implement trend detection using mean, slope, and intercept, with tests for upward, downward, and no trend.

Explore seasonality detection with seasonal differencing in time series, subtracting current values from the previous seasonal period to assess stationarity using period choices like days, weeks, or months.

detect white noise in a time series by computing mean, variance, and autocorrelation, and apply threshold checks to declare presence in a rust-based implementation.

Explore autoregressive time series modeling, where the current value is predicted from past values; learn coefficient estimation, least squares fitting, and forecasting with stationary data.

Learn to build an autoregressive time series model from scratch in Rust by implementing mean, covariance, autocorrelation, fitting coefficients, and making predictions.

Explore the simple moving average and weighted moving average within a window size, using a data vector to compute windowed means and time-step calculations.

Implement a simple moving average in Rust by creating an sma function, computing rolling averages on f64 data with a window size, and collecting results for further use.

Explore the math of weighted moving average on time series data with a three-value window, applying weights and dividing by the sum of weights.

Implement a weighted moving average in Rust by creating a module, computing a weighted sum with weights, and returning a result vector for a given data window.

Discover how arma model combines autoregressive and moving average components to handle stationary data with short-term dependencies, including model order, fitting, evaluation, and forecasting.

Build an ARMA model from scratch in Rust by implementing AR and MA components, generating deterministic white noise, and validating results with cargo run.

Explore autoregressive integrated moving average models, with p, q, and d. Learn how AR and MA components combine with the error term and white noise in ARIMA forecasts.

Implement the arima model from scratch in rust by building an arima struct, handling autoregressive and moving average coefficients, p, d, q, differencing, inverse differencing, and prediction.