A library of time series programs (STB-17: sts7) --------------------------------- The Stata time series library is a group of interconnected ado-files for time series analysis. These programs are continually being updated and are in various stages of completion and reliability. In the list of programs below, each program's status is indicated by a letter grade. These grades have the following interpretation: Status Interpretation ------ -------------- A The program is safe for general use and has been documented in the Stata Technical Bulletin (STB). B The program produces accurate results. However, the program may not be adequately documented, there may be deficiencies in the user or interprogram interface, or the program may not meet all the requirements of an estimation command. This program is likely to be updated soon. C This program is incomplete in significant ways. However, an advanced Stata user (a user who is fluent in reading ado-files and has some experience in writing them) may be able to use this program successfully with some modifications or work-arounds. D This program has serious deficiencies. Advanced Stata users may, nonetheless, wish to review this program for ideas on designing their own ado-files to provide this function. O Obsolete. This program has been superseded by newer code. This grade is typically assigned to old utilities that are still used by one or two user-level programs. User-level programs ------------------- The following table lists the user-level programs in the time series library along with their current status. Separate on-line help is available for each of these programs. Program Status Description ------- ------ ----------- ac A display a correlogram chow C Chow and Farley-Hinich-McGuire tests for a shift in regression coefficients coint B Engle-Granger test for cointegration csunits A set cross-section indexes for a panel data set datevars A set date variables for a time series data set date2obs A return observation number corresponding to a date dickey B Dickey-Fuller test for a unit root dif A generate differences downame A converts day-of-the-week code to a name dropoper A drop operator variables faketemp A create temporary variable names that can be lagged safely filldate A fill in missing date values in projection period finddate A display dates of data availability findlag B find optimal lag length findsmpl B find sample coverage growth A generate growth rate growthi A immediate form of growth hansen A Hansen's test for parameter instability lag A generate lags lastday A calculates the last day of the month lead A generate leads lstbday A calculates the last business day of the month mdytodow A calculates day-of-the-week from month/day/year mnthname A converts month code to a name namedow A converts a name to a day-of-the-week code namemnth A converts a name to a month code pac A display a partial autocorrelation plot pearson A Pearson correlation with p-value period A set period of time series data ppunit B Phillips-Perron test for a unit root projdate A specify initial projection date quandt B calculate Quandt statistics for a break in a regression regdiag B calculate regression diagnostics scenario A generate ad hoc scenario for RHS variable spear A Spearman correlation with p-value tauprob A p-values for unit root and cointegration tests testsum B test the sum of a set of regression coefficients today A calculates today tsfit A estimate a time series regression tsload B load an ad hoc time series equation into memory tsmult A calculate dynamic multipliers after a time series regression tspred A dynamically forecast or simulate a time series regression tsreg A estimate a time series regression, display dynamic multipliers, and calculate diagnostic tests unitroot B unit root tests xcorr A cross correlations ystrday A calculates yesterday from today Utility programs ---------------- The time series library is built on a core of common utility programs. These programs are useful to advanced Stata users who wish to write their own time series programs. These utilities include: Program Status Description ------- ------ ----------- _ac B calculate autocorrelations, standard errors, and Q-statistics _addl B "add" a lag operator to a variable name _addop B "add" an arbitrary operator to a variable name _getrres B calculate recursive residuals for a regression model _gtrank B provide "trank" for the egen command, i.e., calculate ranks over non-missing observations respecting "if" and "in" and without sorting the data _inlist B determine whether a token appears in a token list _invlist B determine whether a varname appears in a varlist _opnum B decode the operators and their powers in a varname _parsevl B parse a varlist to replace abbreviations _partset B partition two sets of tokens into their intersection and complements _sfran B replacement for sfrancia that preserves the data order _subchar B replace one character in a string with another _ts_dsmp B display sample coverage for estimation command _ts_flag B calculate lag length statistics _ts_gdat B format a date value for a given observation _ts_meqn B parse a time series command and generate lags _ts_pars B parse a time series command into useful macros _ts_peri B obtain periodicity of data Support ------- Sean Becketti, Stata Technical Bulletin FAX (913)-888-6708 Also see -------- STB: dm6 (STB-5) ; dm20 (STB-20); ip5 (STB-17); sg5.1 (STB-13); sts1 (STB-5) ; sts2 (STB-7) ; sts3 (STB-13); sts4 (STB-15); sts4 (STB-15); sts6 (STB-17); sts7 (STB-17); sts8 (STB-20); On-line: ^help^ for user-level programs listed above