|
|
At Mark 6, the NAG C Library is offering extended coverage in a number of areas, primarily optimization and statistics.
All functions in the NAG C Library conform fully to ANSI C, and functions introduced at this mark take advantage of the const keyword to allow for safer code which can be more easily optimised. Although existing functions have not yet been modified, it is our intention to extend the use of const and other features of the C99 standard in future releases.
The nonlinear optimizer, nag_opt_nlp (e04ucc), is perhaps the most popular routine in the NAG C Library, with particular use in portfolio management. At Mark 6, we are offering nag_opt_nlp_sparse (e04ugc), which provides the same functionality as nag_opt_nlp (e04ucc) but is for use with large sparse problems.
A set of Time Series routines have been introduced to generate, forecast and estimate Generalized Autoregressive Conditional Heteroskedastic (GARCH) processes. These routines can model univariate regression-GARCH processes of any order and also take into account various types of asymmetry. One of the main applications of GARCH at present is the modelling of financial data with time varying volatility.
In addition we have added a full suite of functions on non-parametric statistics in Chapter g08. Non-parametric statistical analysis is particularly useful when assumptions cannot be made about the nature of the population on which inference is made.
Functions have also been added to Chapters g01 (Simple Calculations on Statistical Data), g02 (Correlation and Regression Analysis), g04 (Analysis of Variance), g11 (Contingency Table Analysis) and g12 (Survival Analysis).
Finally, in cooperation with CERN, we have added a number of special functions to Chapter s, and polynomial solvers for cubic and quartic equations to Chapter c02.
Several function documents have been revised at the mark. In some cases (Chapters d01, e01 and e02) this is to further clarify the use of NAG's memory handling scheme. In other cases we have removed constructions for pre-ANSI compilers or made changes to the examples to provide clearer illustrations of use.
We are also in the process of revising the style of the NAG C Library documentation. The major change is that each of the parameters has been numbered and its type stated explicitly. This change has been applied to new and significantly revised documentation at this mark. The changes, we believe, aid clarity but do not impact in any way on the usage; therefore existing documents will be modified only when other revisions are applied.
All new example programs at Mark 6 are also in a new style, showing typical C language usage; for example, array storage is now allocated dynamically.
| c02akc |
nag_cubic_roots
Zeros of a cubic polynomial with real coefficients |
| c02alc |
nag_quartic_roots
Zeros of a quartic polynomial with real coefficients |
| e04ugc |
nag_opt_nlp_sparse
NLP problem (sparse) |
| g01aec |
nag_frequency_table
Frequency table from raw data |
| g01gbc |
nag_prob_non_central_students_t
Computes probabilities for the non-central Student's t-distribution |
| g01gcc |
nag_prob_non_central_chi_sq
Computes probabilities for the non-central chi2 distribution |
| g01gdc |
nag_prob_non_central_f_dist
Computes probabilities for the non-central F-distribution |
| g01gec |
nag_prob_non_central_beta_dist
Computes probabilities for the non-central beta distribution |
| g01hbc |
nag_multi_normal
Computes probabilities for the multivariate Normal distribution |
| g02byc |
nag_partial_corr
Computes partial correlation/variance-covariance matrix from correlation/variance-covariance matrix computed by nag_corr_cov (g02bxc) |
| g04bcc |
nag_anova_row_col
Analysis of variance, general row and column design, treatment means and standard errors |
| g04dbc |
nag_anova_confid_interval
Computes confidence intervals for differences between means computed by nag_anova_random (g04bbc) or nag_anova_row_col (g04bcc) |
| g04eac |
nag_dummy_vars
Computes orthogonal polynomials or dummy variables for factor/classification variable |
| g05hkc |
nag_generate_agarchI
Univariate time series, generate n terms of either a symmetric GARCH process or a GARCH process with asymmetry of the form (epsilont-1 + gamma)2 |
| g05hlc |
nag_generate_agarchII
Univariate time series, generate n terms of a GARCH process with asymmetry of the form (|epsilont-1| + gamma epsilont-1)2 |
| g05hmc |
nag_generate_garchGJR
Univariate time series, generate n terms of an asymmetric Glosten, Jagannathan and Runkle (GJR) GARCH process |
| g08aac |
nag_sign_test
Sign test on two paired samples |
| g08acc |
nag_median_test
Median test on two samples of unequal size |
| g08aec |
nag_friedman_test
Friedman two-way analysis of variance on k matched samples |
| g08afc |
nag_kruskal_wallis_test
Kruskal–Wallis one-way analysis of variance on k samples of unequal size |
| g08agc |
nag_wilcoxon_test
Performs the Wilcoxon one-sample (matched pairs) signed rank test |
| g08amc |
nag_mann_whitney
Performs the Mann–Whitney U test on two independent samples |
| g08cbc |
nag_1_sample_ks_test
Performs the one-sample Kolmogorov–Smirnov test for standard distributions |
| g08cdc |
nag_2_sample_ks_test
Performs the two-sample Kolmogorov–Smirnov test |
| g08cgc |
nag_chi_sq_goodness_of_fit_test
Performs the chi2 goodness of fit test, for standard continuous distributions |
| g08eac |
nag_runs_test
Performs the runs up or runs down test for randomness |
| g08ebc |
nag_pairs_test
Performs the pairs (serial) test for randomness |
| g08ecc |
nag_triplets_test
Performs the triplets test for randomness |
| g08edc |
nag_gaps_test
Performs the gaps test for randomness |
| g10abc |
nag_smooth_spline_fit
Fit cubic smoothing spline, smoothing parameter given |
| g10acc |
nag_smooth_spline_estim
Fit cubic smoothing spline, smoothing parameter estimated |
| g10bac |
nag_kernel_density_estim
Kernel density estimate using Gaussian kernel |
| g10zac |
nag_order_data
Reorder data to give ordered distinct observations |
| g11bac |
nag_tabulate_stats
Computes multiway table from set of classification factors using selected statistic |
| g11bbc |
nag_tabulate_percentile
Computes multiway table from set of classification factors using given percentile/quantile |
| g12bac |
nag_surviv_cox_model
Fits Cox's proportional hazard model |
| g13asc |
nag_tsa_resid_corr
Univariate time series, diagnostic checking of residuals, following nag_tsa_multi_inp_model_estim (g13bec) |
| g13fac |
nag_estimate_agarchI
Univariate time series, parameter estimation for either a symmetric GARCH process or a GARCH process with asymmetry of the form (epsilont-1 + gamma)2 |
| g13fbc |
nag_forecast_agarchI
Univariate time series, forecast function for either a symmetric GARCH process or a GARCH process with asymmetry of the form (epsilont-1 + gamma)2 |
| g13fcc |
nag_estimate_agarchII
Univariate time series, parameter estimation for a GARCH process with asymmetry of the form (|epsilont-1| + gamma epsilont-1)2 |
| g13fdc |
nag_forecast_agarchII
Univariate time series, forecast function for a GARCH process with asymmetry of the form (|epsilont-1| + gamma epsilont-1)2 |
| g13fec |
nag_estimate_garchGJR
Univariate time series, parameter estimation for an asymmetric Glosten, Jagannathan and Runkle (GJR) GARCH process |
| g13ffc |
nag_forecast_garchGJR
Univariate time series, forecast function for an asymmetric Glosten, Jagannathan and Runkle (GJR) GARCH process |
| s14aec |
nag_real_polygamma
Derivative of the psi function psi (x) |
| s14afc |
nag_complex_polygamma
Derivative of the psi function psi (z) |
| s17alc |
nag_bessel_zeros
Zeros of Bessel functions Jalpha(x), J'alpha(x), Yalpha(x) or Y'alpha(x) |
| s18ecc |
nag_bessel_i_nu_scaled
Scaled modified Bessel function e-x Inu/4 (x) |
| s18edc |
nag_bessel_k_nu_scaled
Scaled modified Bessel function ex Knu/4 (x) |
| s18eec |
nag_bessel_i_nu
Modified Bessel function Inu/4 (x) |
| s18efc |
nag_bessel_k_nu
Modified Bessel function Knu/4 (x) |
| s18egc |
nag_bessel_k_alpha
Modified Bessel functions Kalpha+n (x) for real x > 0, selected values of alpha ≥ 0 and n = 0,1,...,N |
| s18ehc |
nag_bessel_k_alpha_scaled
Scaled modified Bessel functions functions Kalpha+n (x) for real x > 0, selected values of alpha ≥ 0 and n = 0,1,...,N |
| s18ejc |
nag_bessel_i_alpha
Modified Bessel functions Ialpha +n-1 (x) or Ialpha -n+1 (x) for real x ≠ 0, non-negative alpha < 1 and n = 1,2,...,|N|+1 |
| s18ekc |
nag_bessel_j_alpha
Bessel functions Jalpha +n-1 (x) or Jalpha -n+1 (x) for real x ≠ 0, non-negative alpha < 1 and n = 1,2,...,|N|+1 |
| s21cbc |
nag_jacobian_elliptic
Jacobian elliptic functions sn, cn and dn with complex arguments |
| s21ccc |
nag_jacobian_theta
Jacobian theta functions with real arguments |
| s21dac |
nag_elliptic_integral_f
Elliptic integrals of the second kind with complex arguments |
| s22aac |
nag_legendre_p
Legendre and associated Legendre functions of the first kind with real arguments |
The functions listed below are scheduled for withdrawal from the NAG C Library, because improved functions have now been included in the Library. Users are advised to stop using functions which are scheduled for withdrawal and to use recommended replacement functions instead.
The following functions will be withdrawn at Mark 8.
|
