Vector Statistics & Data Fitting
Speed Statistical Functions Performance
Accelerate operations for applications in computational finance, life sciences, engineering simulations, databases, and other areas. With improved Monte Carlo application performance, deliver random number generators, probability distributions, and capabilities for parallel computation. Pair random-number generators such as Mersenne Twister and Niederreiter with a variety of probability distributions, including uniform, Gaussian, and exponential.
Get computationally intensive building blocks that analyze core statistics both in- and out-of-core to efficiently cover cases when the dataset is too large to fit in the memory all at once. These statistical functions include:
- Computing basic statistics
- Estimating dependencies
- Detecting data outliers
- Replacing missing values
Linear convolution and correlation transformations provide real and complex data with single and double precision while splines for data-fitting capabilities accelerate one-dimensional interpolation for histograms, geometric modeling, surface approximation, and more. The included splines are: linear, quadratic, cubic, lookup, basis (B-splines), monotonic, stepwise, constant, and user-defined.