{"id":22225466,"url":"https://github.com/beliavsky/starpac","last_synced_at":"2026-01-06T06:04:59.169Z","repository":{"id":260771792,"uuid":"876719901","full_name":"Beliavsky/STARPAC","owner":"Beliavsky","description":"Standards Time Series and Regression Package, a library of Fortran subroutines for statistical data analysis developed by the Statistical Engineering Division of the National Institute of Standards and Technology","archived":false,"fork":false,"pushed_at":"2024-11-02T13:12:54.000Z","size":1151,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2025-01-30T07:42:41.080Z","etag":null,"topics":["arima","autocorrelation","fft","forecasting","gamma-function","nonlinear-least-squares","periodogram","polynomial-regression","rng","statistics","time-series-analysis"],"latest_commit_sha":null,"homepage":"","language":"Fortran","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Beliavsky.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-10-22T12:54:37.000Z","updated_at":"2024-11-02T13:23:04.000Z","dependencies_parsed_at":"2024-11-02T14:17:49.144Z","dependency_job_id":"dfe98434-344b-4c07-af29-f4caaf592a91","html_url":"https://github.com/Beliavsky/STARPAC","commit_stats":null,"previous_names":["beliavsky/starpac"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Beliavsky%2FSTARPAC","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Beliavsky%2FSTARPAC/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Beliavsky%2FSTARPAC/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Beliavsky%2FSTARPAC/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Beliavsky","download_url":"https://codeload.github.com/Beliavsky/STARPAC/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":245423887,"owners_count":20612859,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["arima","autocorrelation","fft","forecasting","gamma-function","nonlinear-least-squares","periodogram","polynomial-regression","rng","statistics","time-series-analysis"],"created_at":"2024-12-03T00:17:56.290Z","updated_at":"2026-01-06T06:04:59.115Z","avatar_url":"https://github.com/Beliavsky.png","language":"Fortran","funding_links":[],"categories":[],"sub_categories":[],"readme":"# STARPAC\n[Standards Time Series and Regression Package](https://water.usgs.gov/software/OTIS/addl/starpac/), a library of Fortran subroutines for statistical data analysis developed by the Statistical Engineering Division of the National Institute of Standards and Technology by Janet R. Donaldson and Peter V. Tryon, converted to free source form by [John Burkardt](https://people.math.sc.edu/Burkardt/f_src/starpac/starpac.html), who supplied the list of routines below. Compile with `gfortran starpac.f90 starpac_test.f90`. Also compiles with ifort and g95.\n\n```\nList of Routines:\n\nABSCOM counts the entries of | V(1:N) - W(1:N) | greater than ABSTOL.\nACCDIG returns the number of accurate digits in an approximation to X.\nACFD computes autocorrelations and partial autocorrelations.\nACFDTL prints titles for ACORRD.\nACFER does error checking for the ACF routines.\nACF is the simple interface to the autocorrelations routines.\nACFF computes autocorrelations of a time series using an FFT.\nACFFS uses an FFT with ACVF estimates for autocorrelations of a time series.\nACFLST lists the autocorrelations and other information.\nACFM computes autocorrelations of a time series with missing data.\nACFMN computes autocorrelations of a time series.\nACFMNF computes autocorrelations of a time series.\nACFMNM computes autocorrelations of a time series with missing data.\nACFMS is the user interface for autocorrelations of a time series with missing data.\nACFOUT prints autocorrelations.\nACFSD computes the standard error of autocorrelations.\nACFSDM computes the standard error of autocorrelations with missing data.\nACFS computes autocorrelations with computed ACVF estimates.\nACVF computes the autocovariance function of a series.\nACVFF computes the ACVF of a series using two FFT passes.\nACVFM computes autocovariance when missing data is involved.\nADJLMT corrects the plot limits when all observations are equal.\nAIMEC is the user interface for ARIMA estimation.\nAIME is the user interface for ARIMA estimation, control call.\nAIMES is the user interface for ARIMA estimation, long call.\nAIMF is the user interface for ARIMA estimation, short call.\nAIMFS is the user interface for ARIMA estimation, control call.\nAIMX1 sets the starting parameter values for AIMX.\nALBETA computes the logarithm of the Beta function.\nALGAMS evaluates the log of the absolute value of the Gamma function.\nALNGAM computes the logarithm of the absolute value of the Gamma function.\nALNREL evaluates log ( 1 + X ) with relative error control.\nAMDRV estimates the jacobian matrix.\nAMEAN computes the arithmetic mean of a series.\nAMEANM computes the arithmetic mean of a series with missing data.\nAMECNT is the control routine for nonlinear least squares regression.\nAMEDRV is the control routine for nonlinear least squares regression.\nAMEER checks errors for the nonlinear least squares estimation.\nAMEFIN analyzes nonlinear least squares estimates after they are computed.\nAMEHDR prints headings for nonlinear least squares estimation.\nAMEISM prints an initial summary for nonlinear least squares routines.\nAMEMN is the control routine for using the NL2 software package.\nAMEOUT prints the final summary output from ARIMA estimation.\nAMEPT1 prints data summary for nonlinear least squares routines.\nAMEPT2 prints four standardized residual plots.\nAMESTP controls the step size selection.\nAMFCNT is the control routine for ARIMA forecasting.\nAMFER checks errors for nonlinear least squares estimation.\nAMFHDR prints headers for nonlinear least squares estimation.\nAMFMN computes and prints ARIMA forecasts.\nAMFOUT produces ARIMA forecasting output.\nAMLST1 prints parameters for the ARIMA routine.\nAMLST prints parameter summaries from ARIMA forecasting.\nAOS computes autoregressive model order selection statistics.\nAOSLST lists the autoregressive model order selection statistics.\nAOV1ER does preliminary checks on input to the one-way family.\nAOV1 is a user interface to AOV1MN, one-way analysis of variance.\nAOV1HD prints headers for the one-way ANOVA family.\nAOV1MN computes results for analysis of a one-way classification.\nAOV1S is a user interface for AOV1MN, one-way analysis of variance.\nAOV1XP prints storage for one-way family exerciser.\nARCOEF uses Durbin's method for autoregression coefficients with order lag.\nARFLT performs autoregressive filtering.\nASSESS assesses a candidate step.\nAXPBY: SZ(1:N) = SA * SX(1:N) + SB * SY(1:N).\nBACKOP computes the number of back order terms for an ARIMA model.\nBETAI computes the incomplete Beta ratio.\nBFSDRV is the driver for time series Fourier spectrum analysis.\nBFSER checks errors for time series Fourier univariate spectrum analysis.\nBFS: short interface for time series Fourier bivariate spectrum analysis.\nBFSF: short interface for time series Fourier bivariate spectrum analysis.\nBFSFS: long interface for time series Fourier bivariate spectrum analysis.\nBFSLAG: lag window truncation points for Fourier bivariate spectral analysis.\nBFSM: short interface for time series bivariate Fourier spectrum analysis.\nBFSMN computes square coherency and phase components of a bivariate spectrum.\nBFSMS: long interface for BFS analysis with missing observations.\nBFSMV: short interface for BFS analysis, missing observations, covariances.\nBFSMVS: long interface for BFS analysis, missing observations, covariances.\nBFSS: long call for time series bivariate Fourier spectrum analysis.\nBFSV: short call for BFS analysis, with covariance input rather than series.\nBFSVS: long call for BFS analsys with covariances input rather than series.\nCCFER does error checking for CCF routines.\nCCF computes the cross-correlation of two time series.\nCCFF computes the cross-correlation of two time series by Singleton's FFT.\nCCFFS computes multivariate cross-correlations and covariances by FFT.\nCCFLST lists cross-correlations, standard errors, and summary information.\nCCFM computes cross-correlation of two series with missing data.\nCCFMN is the main routine for cross-correlations.\nCCFMNF is the main routine for cross-correlations using an FFT.\nCCFMNM is the main routine for cross-correlations with missing data.\nCCFMS is a user routine for multivariate cross-correlations.\nCCFOUT prints cross-correlations and standard errors.\nCCFSD is the main routine for computing standard error of cross-correlations.\nCCFSDM: standard error of cross-correlations with missing data.\nCCFS is the user routine for multivariate cross-correlations.\nCCFXP lists results for the time series cross-correlation routines.\nCCVF computes the cross covariance function between two series.\nCCVFF computes the cross covariance function between two series.\nCCVFM computes the cross covariance function of two series with missing data.\nCDFCHI computes the CDF for the Chi Square distribution.\nCDFF computes the CDF for the F distribution.\nCDFNML computes the CDF for the standard normal distribution.\nCDFT computes the CDF for Student's T distribution.\nCENTER centers an observed series.\nCHIRHO computes the Chi Square statistic and its probability.\nCMPFD computes a finite difference derivative.\nCNTR centers the input seriers about its mean.\nCORRER checks for errors in the input parameters.\nCORR: short call to correlation family of routines.\nCORRHD prints headers for the correlation family.\nCORRMN is the main routine in the correlation family.\nCORRS is the user routine for the correlation, with a long call interface.\nCORRXP prints stored output returned from CORRS.\nCOVCLC computes the covariance matrix for NL2ITR.\nCPYASF copies a symmetric matrix stored rowwise into rectangular storage.\nCPYMSS copies an N by M matrix.\nCPYVII copies an integer vector.\nCSEVL evaluates a Chebyshev series.\nD1MACH returns double precision machine constants.\nDCKCNT controls the derivative checking process.\nDCKCRV checks whether high curvature caused poor derivative approximation.\nDCKDRV is the driver to the derivative checking routines.\nDCKER does error checking for the derivative checking routines.\nDCKFPA checks if arithmetic precision causes poor derivative approximation.\nDCKHDR prints page headers for the derivative checking routines.\nDCKLS1 sets up a problem for testing the step size selection family.\nDCKLSC is the user routine for comparing analytic and numeric derivatives.\nDCKLS is the user routine for comparing analytic and numeric derivatives.\nDCKMN is the main routine for checking analytic versus numeric derivatives.\nDCKOUT prints results from the derivative checking routine.\nDCKZRO rechecks derivative errors where the analytic derivative is zero.\nDCOEF expands a difference filter.\nDEMDRV is the driver routine to demodulate a series.\nDEMOD demodulates a series at a given frequency.\nDEMODS demodulates a series at a given frequency.\nDEMODU demodulates a series at a given frequency.\nDEMORD sets up the data for the phase plots.\nDEMOUT prints output for the time series demodulation routines.\nDFAULT supplies default values to IV and V.\nDFBW computes degrees of freedom and bandwidth for a given lag window.\nDFBWM computes DOF and BW for a given lag window with missing data.\nDIFC expands a difference filter and performs difference filtering.\nDIF performs a first difference filtering operation.\nDIFMC expands a difference filter and performs the difference filter.\nDIFM performs a first difference filter for a series with missing data.\nDIFSER performs a differencing operation on a series.\nDOTC computes the dot product of two series, centered about their means.\nDOTCM computes the dot product of series with missing data.\nDOTPRD returns the inner product of two vectors.\nDRV1A derivative function for NLS family exerciser subroutine MDL1.\nDRV1B is an INCORRECT derivative function for the NLS exerciser MDL1.\nDRV2 is a derivative function for the NLS exerciser routine MD12.\nDRV3 is the derivative function for NLS family exerciser subroutine MDL3.\nDRV4A is a (correct) derivative for testing derivative checking routines.\nDRV4B is an (incorrect) derivative for testing derivative checking routines.\nDRV is a dummy derivative function for the NLS family.\nDUPDAT updates the scale vector for NL2ITR.\nE9RINT stores the current error message or prints the old one.\nECVF prints an error message if missing data affects the covariance lags.\nEHDR prints the heading for the error checking routines.\nEIAGE ensures that \"not too many\" vectors are below a given lower bound.\nEIAGEP prints the error messages for ERAGT and ERAGTM.\nEISEQ prints an error message if NVAL is not equal to NEQ.\nEISGE prints a warning if NVAL is less than NMIN.\nEISII warns if an integer value does not lie within a given range.\nEISLE warns if an integer is greater than a given maximum.\nEISRNG warns if ISEED is not a suitable random number seed.\nEIVEO checks whether all vector entries are even (or odd).\nEIVEQ warns if the vector does not have at least NEQMN entries equal to IVAL.\nEIVII warns if too many values are outside given limits.\nENFFT checks that NFFT is suitable for the Singleton FFT routine.\nEPRINT prints the last error message, if any.\nERAGT warns if too many values are less than a lower bound.\nERAGTM warns if too many values are less than or equal to a lower bound.\nERAGTP prints the error messages for ERAGT and ERAGTM.\nERDF checks the values that specify differencing on a time series.\nERFC evaluates the complementary error function.\nERF evaluates the error function.\nERIODD warns if the value of NVAL is inconsistent.\nERSEI warns if a value is not between given limits.\nERSGE warns if a value is not greater than or equal to a minimum value.\nERSGT warns if the input value is not greater than a specified minumum.\nERSIE warns if a value is not within a specified range.\nERSII warns if the input value is not within the given range.\nERSLF checks the definition of a symmetric linear filter.\nERSLFS checks values specifying a symmetric linear filter for a time series.\nERVGT ensures that \"most\" values are greater than a specified lower bound.\nERVGTM ensures that \"most\" values are greater than a specified lower bound.\nERVGTP prints the error messages for ERVGT and ERVGTM.\nERVII checks for vector values outside given limits.\nERVWT checks user-supplied weights.\nETAMDL computes noise and number of good digits in model routine results.\nEXTEND returns the I-th term in a series.\nFACTOR factors an integer.\nFDUMP is a dummy version of the dump routine called by XERRWV.\nFFTCT does a cosine transform of n=2*n2 symmetric data points.\nFFT is a multivariate complex Fourier transform.\nFFTLEN computes the value of NFFT for the Singleton FFT routine.\nFFTR is the user-callable routine for the Fourier transform of a series.\nFITEXT checks whether the fit is exact to machine precision.\nFITPT1 prints the data summary for nonlinear least squares routines.\nFITPT2 prints the four standardized residual plots.\nFITSXP generates reporst for least squares exerciser returned storage.\nFITXSP generates reports for least squares exerciser returned storage.\nFIXPRT sets the character array 'FIXED'.\nFLTAR filters an input series using an autoregressive filter.\nFLTARM filters a series with missing data, using an autoregressive filter.\nFLTMA filters a series using a simple moving average filter.\nFLTMD applies a modified Daniel filter to a symmetric series.\nFTLSL filters an input series.\nGAMI evaluates the incomplete Gamma function.\nGAMIT evaluates Tricomi's incomplete Gamma function.\nGAMLIM calculates the legal range of input arguments for the Gamma function.\nGAMMA evaluates the Gamma function.\nGAMR computes the reciprocal of the Gamma function.\nGENI assigns an arithmetic sequence of values into an integer vector.\nGENR puts an arithmetic sequence of values into a real vector.\nGETPI returns the value of PI.\nGFAEST computes the gain of an autoregressive linear filter.\nGFARF: short call to compute gain function of autoregressive filter.\nGFARFS: short call to compute gain function of autoregressive filter.\nGFORD produces ordinants for the gain function plots.\nGFOUT produces the gain function plots.\nGFSEST: gain function of symmetric linear filter with given frequencies.\nGFSLF: short call for gain function of symmetric linear filter.\nGFSLFS: short call for gain function of symmetric linear filter.\nGMEAN computes the geometric mean of a series.\nGQTSTP computes the Goldfeld-Quandt-Trotter step by More-Hebden technique.\nHIPASS carries out a high-pass filtering of a series.\nHISTC: long call for producing a histogram.\nHIST: short call for producing a histogram.\nHPCOEF computes hi-pass filter given K-term low pass filter coefficients.\nHPFLT compute high pass filter coefficients corresponding to a low pass filter.\nHSTER does error checking for the HIST family of histogram routines.\nHSTMN is the main routine for producing histograms.\nI1MACH returns integer machine constants.\nI8SAVE returns the current error number or recovery switch.\nICNTI counts the number of occurences of a value in an integer vector.\nICOPY copies an integer vector.\nIMDCON returns integer machine-dependent constants.\nINITS initializes an orthogonal series.\nINPERL computes the number of elements that can be printed on one line.\nIPGDV produces coordinates for the spectral plots.\nIPGM: short call to compute the integrated periodogram of a series.\nIPGMN computes the integrated periodogram.\nIPGMP is the user routine for integrated periodograms of a series (short call).\nIPGMPS is the user routine for the integrated periodogram of a series (long call).\nIPGMS is the user routine fo the integrated periodogram of a series (long call).\nIPGORD produces coordinates for the spectral plots.\nIPGOUT produces the integrated periodogram plots.\nIPRINT sets the logical unit for printed output.\nISAMAX indexes the real array element of maximum absolute value.\nITSMRY prints an iteration summary.\nJ4SAVE saves and recalls data needed by the XERROR error library.\nLDSCMP computes storage needed for arrays.\nLINVRT computes the inverse of a lower triangular matrix.\nLITVMU solves L' * X = Y, where L is a lower triangular matrix.\nLIVMUL solves L * X = Y, where L is a lower triangular matrix.\nLLCNT is the controlling subroutine for linear least squares.\nLLCNTG is the controlling subroutine for general linear least squares.\nLLCNTP is the controlling subroutine for polynomial linear least squares.\nLLER is the error checking routine for the linear least squares routines.\nLLHDRG: page headings for the unrestricted linear least squares routines.\nLLHDRP: page headings for polynomial linear least squares routines.\nLLS is the general linear model least squares fit routine.\nLLSMN: main program for the linear least squares fitting.\nLLSP does an unweighted polynomial model least squares fit.\nLLSPS does an unweighted polynomial model least squares fit.\nLLSPW does a weighted polynomial model least squares fit.\nLLSPWS computes a weighted polynomial model least squares fit.\nLLSS computes an unweighted linear model least squares fit.\nLLSW computes a weighted linear model least squares fit.\nLLSWS performs a general linear model weighted least squares fit.\nLMSTEP computes a Levenberg-Marquardt step by More-Hebden techniques.\nLOGLMT adjusts plot limits for log plots, and computes log axis labels.\nLOPASS carries out a low-pass filtering of a series.\nLPCOEF computes a least squares approximation to an ideal low pass filter.\nLPFLT computes the low-pass filter coefficients.\nLSAME returns TRUE if CA is the same letter as CB regardless of case.\nLSQRT computes the Cholesky factor of a lower triangular matrix.\nLSTLAG finds the last computable lag value.\nLSTVCF prints N elements of a masked array.\nLSTVEC prints indices and values of a real vector.\nLSVMIN estimates the smallest singular value of a lower triangular matrix.\nLTSQAR sets A to the lower triangle of L' * L.\nMADJ is a sample jacobian routine.\nMADR is a sample residual routine.\nMAFLT performs a moving average filtering operation.\nMATPRF prints a square matrix stored in symmetric form.\nMATPRT is a matrix printing routine.\nMDFLT is a user routine for a modified Daniels filter of symmetric series.\nMDL1 is the model function for an NLS exerciser.\nMDL2 is a model function for an NLS exerciser.\nMDL3 is a model function for an NLS exerciser.\nMDL4 is a model routine for step size and derivative checking routines.\nMDLTS1 is the user callable routine for estimating box-jenkins arima models.\nMDLTS2 is the model routine for Pack's specification of box-jenkins models.\nMDLTS3 is the user callable routine for estimating box-jenkins arima models.\nMGS solves a linear system using modified Gram-Schmidt algorithm.\nMODSUM prints the model summary for the ARIMA routines.\nMPPC produces a simple page plot with multiple Y-axis values.\nMPP produces a simple page plot with multiple Y-axis values.\nMPPL produces a simple page plot with multiple Y-axis values, and log option.\nMPPMC: produce a page plot with multiply Y-axis values, and missing data.\nMPPM: produce a page plot with multiple Y-axis values and missing data.\nMPPML: plot multiple Y-axis values with missing data, log option.\nMSGX prints the returned and expected values for the error flag IERR.\nMULTBP multiplies two difference factors from a Box-Jenkins time series model.\nMVCHK checks whether the input value equals the flag value for missing data.\nMVPC produces a vertical plot with multiple Y-axis values.\nMVP produces a vertical plot with multiple Y-axis values.\nMVPL produces a vertical plot with multiple y-axis values (log plot option).\nMVPMC: vertical plot with missing data and multiple y-axis values (long call).\nMVPM: vertical plot with missing data and multiple y-axis values (short call).\nMVPML: vertical plot with missing data and multiple y-axis values (log option).\nNCHOSE combines difference factors from a Box-Jenkins time series model.\nNL2ITR carries out iterations for NL2SOL.\nNL2SNO is like NL2SOL, but uses a finite difference jacobian.\nNL2SOL minimizes a nonlinear sum of squares using an analytic jacobian.\nNL2X tests nl2sol and nl2sno on madsen example.\nNLCMP computes statistics for the NLS family when weights are involved.\nNLCNTA: controlling routine for NLS regression with analytic derivatives.\nNLCNT controlling subroutine for nonlinear least squares regression.\nNLCNTN controlling routine for NLS regression with approximate derivatives.\nNLDRVA computes the analytic derivative matrix from the user DERIV routine.\nNLDRVN approximates the derivative matrix.\nNLER does error checking routine for nonlinear least squares estimation.\nNLERR sets the error flag ierr based on the convergence code returned by NL2.\nNLFIN completes the NLS analysis once the estimates have been found.\nNLHDRA prints headings for NLS estimation using analytic derivatives.\nNLHDRN prints headings for NLS estimation using approximate derivatives.\nNLINIT initializes the NLS routines.\nNLISM prints an initial summary for the nonlinear least squares routines.\nNLITRP prints iteration reports for nonlinear least squares regression.\nNLMN: controlling routine for nonlinear least squares regression.\nNLOUT prints the final summary report for nonlinear least squares routines.\nNLSC: NLS regression, approximate derivatives (control call)\nNLSDC: NLS regression, analytic derivatives, user parameters.\nNLSD: nonlinear least squares regression, analytic derivatives (short call).\nNLSDS: NLS regression, analytic derivatives, user parameters.\nNLS: NLS regression with numeric derivatives, short call.\nNLSKL prints warning messages for the nonlinear least squares routines.\nNLSPK packs the unmasked elements of one vector into another.\nNLSS: interface for nonlinear least squares reqression, approximate derivatives.\nNLSUPK unpacks a vector into another, using a mask vector.\nNLSWC: nonlinear least squares regression with weights and approximate derivatives.\nNLSWDC: NLS regression, analytic derivatives, weights, user parameters.\nNLSWD: NLS regression, analytic derivatives, weights (short call).\nNLSWDS: NLS regression with analytic derivatives, weights, user parameters.\nNLSW: NLS regression with approximate derivatives and weights.\nNLSWS: NLS regression with approximate derivatives and weights.\nNLSX1 sets the starting parameter values for NLSX.\nNLSX2 sets a problem for testing the NLS family.\nNRANDC generates pseudorandom normally distributed values.\nNRAND generates pseudorandom normally distributed values.\nOANOVA computes and prints analysis of variance.\nOBSSM2 lists the data summary for the arima estimation routines.\nOBSSUM lists the data summary for the least squares subroutines.\nPARCHK checks the NL2SOL parameters.\nPARZEN computes and stores the Parzen lag window.\nPGMEST computes the periodogram estimates.\nPGM is the user callable routine for the raw periodogram of a series.\nPGMMN is the main routine for computing the raw periodogram.\nPGMS computes the (raw) periodogram of a series (long call).\nPGORD produces coordinates for the periodogram plot.\nPGOUT produces the periodogram plots.\nPLINE defines one line of a plot string for the vertical plot routines.\nPLTCHK checks for errors for the multiple plot routines.\nPLTPLX computes the point location in the plot string.\nPLTSYM supplies the plot symbol for the plot line.\nPOLAR converts complex numbers from Cartesian to polar representation.\nPPC produces a simple page plot (long call).\nPPCNT is the controling routine for user called page plot routines.\nPP is the user callable routine which produces a simple page plot (short call).\nPPCHFS computes the percentage points of the Chi Square distribution.\nPPFF computes the percentage points for the F distribution.\nPPFNML computes the percentage points of the normal distribution.\nPPFT computes the percentage points of the Student's T distribution.\nPPL produces a simple page plot (log option).\nPPLMT sets the plot limits for page plots with missing values.\nPPMC produces a simple page plot for data with missing observations (long call).\nPPM produces a simple page plot for data with missing observations (short call).\nPPML plots data with missing observations (log option).\nPPMN is the main routine for page plots.\nPRTCNT sets up the print control parameters.\nQAPPLY applies orthogonal transformation to the residual R.\nQRFACT computes the QR decomposition of a matrix.\nR1MACH returns single precision machine constants.\nR9GMIT computes Tricomi's incomplete gamma function for small X.\nR9LGIC compute the log complementary incomplete gamma function.\nR9LGIT computes the log of Tricomi's incomplete Gamma function.\nR9LGMC computes the log Gamma correction factor.\nRANDN returns normal random numbers.\nRANDU returns uniform random numbers.\nRANKO puts the rank of N X's in the vector R.\nREALTR computes the forward or inverse Fourier transform of real data.\nRELCOM computes the difference between V(I) and W(I) relative to RELTOL.\nRELDST computes the relative difference between two real values.\nREPCK reformats the data in D for the N by NPAR format used by NLCMP.\nRMDCON returns machine constants.\nRPTMUL multiplies the R factor times a vector X.\nS88FMT writes an integer into a string.\nSAMPLE creates a new series by sampling every K-th item of the input.\nSASUM takes the sum of the absolute values of a real vector.\nSAXPY adds a real constant times one vector to another.\nSCOPY copies one real vector into another.\nSDOT forms the dot product of two real vectors.\nSETERR sets the error number and prints the error message.\nSETESL computes the smallest suitable value of NFFT for given N and Singleton FFT.\nSETFRQ computes the frequencies at which the spectrum is to be estimated.\nSETIV sets the entries of an integer vector to a value.\nSETLAG sets the number of autocorrelations to be computed.\nSETRA sets the entries of a real array to a given value.\nSETROW selects the row used by the derivative checking procedure.\nSETRV sets the elements of a real vector to a value.\nSLFLT applies a symmetric filter to a series.\nSLUPDT updates a symmetric matrix A so that A * STEP = Y.\nSLVMUL sets Y = S * X, where S is a P by P symmetric matrix.\nSMACH computes machine parameters for single precision arithmetic.\nSMPLY samples every K-th observation from a series.\nSNRM2 computes the Euclidean norm of a real vector.\nSPCCK analyzes ordinates for the spectal semi-log plots.\nSPPC produces a simple page plot with user control of plot symbols (long call).\nSPP produces a simple page plot with user control of plot symbols (short call).\nSPPL produces a simple page plot with user control of plot symbols (log option).\nSPPLTC: confidence interval and bandwidth coordinates for spectrum plots.\nSPTLTD sets various y axis limits for decibel spectrum plots.\nSPPLTL sets various y axis limits for decibel spectrum plots.\nSPPMC: page plot with user plot symbols and missing observations (long call).\nSPPM page plot with user plot symbols and missing observations (short call).\nSPPML: page plot with user plot symbols and missing observations (log option).\nSROT applies a real plane rotation.\nSROTG constructs a real Givens plane rotation.\nSRTIR sorts an integer array IR on a key array A.\nSRTIRR sorts arrays A, IR and RR based on values in A.\nSRTRI sorts array A on integer array IR.\nSRTRRI sorts the array IR, A and RR, based on values in IR.\nSSCAL scales a real vector by a constant.\nSSIDI computes the determinant, inertia and inverse of a real symmetric matrix.\nSSIFA factors a real symmetric matrix.\nSSWAP interchanges two real vectors.\nSTAT1 computes statistics for a sorted vector.\nSTAT1W computes statistics for a sorted vector with weights.\nSTAT2 computes statistics that do not require sorted data.\nSTAT2W computes statistics on an unsorted, weighted vector.\nSTATER does error checking for the STAT family of routines.\nSTAT computes 53 statistics for an unweighted vector.\nSTATS computes 53 different statistics for an unweighted vector.\nSTATW computes 53 statistics for a weighted vector.\nSTATWS computes 53 statistics for a weighted vector.\nSTKCLR clears the stack for framework area manipulation routines.\nSTKGET allocates space on an integer stack.\nSTKREL de-allocates the last allocations made in the stack.\nSTKSET initializes the stack to NITMES of type ITYPE.\nSTKST returns statistics on the state of the stack.\nSTOPX is called to stop execution.\nSTPADJ adjusts the selected step sizes to optimal values.\nSTPAMO is a dummy routine for the arima estimation routines.\nSTPCNT controls the stepsize selection process.\nSTPDRV is the driver for selecting forward difference step sizes.\nSTPER does error checking for the stepsize selection routines.\nSTPHDR prints page headings for the stepsize selection routines.\nSTPLS1 sets a test problem for the step size selection family.\nSTPLS2 sets a test problem for the step size selection family.\nSTPLSC selects step sizes for forward difference estimates of derivatives.\nSTPLS selects step sizes for estimating derivatives in NLS routines.\nSTPMN: main routine for numerical derivative step size selection.\nSTPOUT prints results for the step size selection routines.\nSTPSEL selects new step sizes until no further improvement can be made.\nSTRCO estimates the condition of a real triangular matrix.\nSTRDI computes the determinant and inverse of a real triangular matrix.\nSUMBS finds a zero or value closest to zero in a sorted vector.\nSUMDS sums unweighted powers of differences from the mean of a sorted vector.\nSUMID sums I * ( X(I) - XMEAN ).\nSUMIDW: dot product of I and ( X(I) - XMEANW ).\nSUMOT reports the computation of 53 selected statistics.\nSUMSS calculates the sum of powers and mean for a sorted vector.\nSUMTS calculates unweighted trimmed mean for a sorted vector.\nSUMWDS calculates sums of powers of differences from the weighted mean.\nSUMWSS calculates weighted and unweighted sums of powers and the mean.\nSUMWTS calculates weighted and unweighted means for a sorted vector.\nSVPC produces a vertical plot with user plot symbols (long call).\nSVP: vertical plot with user plot symbols (short call).\nSVPL produces a vertical log plot with user control of the plot symbol.\nSVPMC: vertical plot with missing data and user plot symbols (long call).\nSVPM: vertical plot with missing data and user plot symbols (short call).\nSVPML: vertical plot with missing data and user plot symbols (log plot option).\nTAPER applies a split-cosine-bell taper to a centered observed series.\nTIMESTAMP prints the current YMDHMS date as a time stamp.\nUAS is the user callable routine for autoregressive spectrum estimation.\nUASCFT computes autoregressive coefficients using Durbin's method.\nUASDV is the driver for computing the autoregressive and Fourier spectrums.\nUASER: error checks for time series Fourier univariate spectrum analysis.\nUASET calculates the autoregressive spectrum.\nUASF is the user callable routine for autoregressive spectrum estimation.\nUASFS: interface for autoregressive spectrum estimation using the FFT (long call).\nUASORD produces coordinates for the spectrum plots.\nUASOUT produces the spectrum plots for the autoregressive spectrum estimates.\nUASS: user interface for autoregressive spectrum estimation (long call).\nUASVAR computes the variance for a given series and autoregressive model.\nUASV is the user routine for autoregressive spectrum estimation.\nUASVS is a user routine for autoregressive spectrum estimation.\nUFPARM is a dummy version of the optional user function for NL2SOL.\nUFSDRV is the controlling routine for time series Fourier spectrum analysis.\nUFSET checks errors for time series Fourier univariate spectrum analysis.\nUFSEST computes the spectra and the confidence limits.\nUFS: user routine for time series Fourier spectrum analysis (short call).\nUFSF: user routine for Fourier spectrum analysis using fft (short call).\nUFSFS: user routine for Fourier spectrum analysis using the fft (long call).\nUFSLAG computes the lag window truncation points for spectrum analysis.\nUFSM: user routine for Fourier spectrum analysis with missing data (short call).\nUFSMN computes autocorrelations and partial autocorrelations of a time series.\nUFSMS: time series Fourier spectrum analysis with missing data (long call).\nUFSMV: Fourier spectrum analysis, missing data, user ACVF values (short call).\nUFSMVS: time series Fourier spectrum analysis with missing data (long call).\nUFSOUT produces the Fourier bivariate spectrum output.\nUFSPCV produces coordinates for the spectrum plots.\nUFSS: time series Fourier spectrum analysis (long call).\nUFSV: Fourier spectrum analysis, user supplied ACVF values (short call).\nUFSVS: Fourier spectrum analysis and user supplied acvf values (long call).\nV2NORM computes the L2 norm of a vector.\nVCOPY copies a vector.\nVCVOTF prints the variance-covariance matrix.\nVCVOUT prints the variance-covariance matrix.\nVERSP prints the version number.\nVP is the user callable routine which produces a vertical plot (short call).\nVPC is the user callable routine which produces a vertical plot (long call).\nVPCNT is the controlling routine for user-called vertical plots\nVPHEAD prints the heading for the vertical plot output.\nVPL is the user callable routine which produces a vertical log plot.\nVPLMT sets the plot limits for vertical plots\nVPM produces a vertical plot with missing data (short call).\nVPMC produces a vertical plot with missing data (long call).\nVPML produces a vertical plot with missing data (log plot option).\nVPMN produces vertical plots.\nXERABT aborts execution of the program.\nXERBLA is an error handler for the LAPACK routines.\nXERCLR resets the current error number to zero.\nXERCTL gives the user control over handling individual errors.\nXERPRT prints an error message.\nXERROR processes a diagnostic message.\nXERRWV processes a diagnostic message.\nXERSAV records that an error has occurred.\nXGETF returns the current error control flag.\nXGETUA determines the units to which error messages are being sent.\nXSETF sets the error control flag.\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbeliavsky%2Fstarpac","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbeliavsky%2Fstarpac","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbeliavsky%2Fstarpac/lists"}