To declare a pool object, use the pool keyword, followed by a pool name, and optionally, a list of pool members. Pool members are short text identifiers for the cross section units:

To read data using the pool object:

To delete and store pool series you may enter:

Descriptive statistics may be computed using the command:

To estimate a pool equation using least squares and to access the t-statistics, enter:

List the cross-section identifiers to add to the pool.

Adds US and GR as cross-section members of the pool object COUNTRIES.

See “Cross-section Identifiers” for a discussion of pool identifiers.

See also Pool::drop.

The cellipse view displays confidence ellipses for pairs of coefficient restrictions for an estimation from a pool object.

Enter the object name, followed by a period, and the keyword cellipse. This should be followed by a list of the coefficient restrictions. Joint (multiple) coefficient restrictions should be separated by commas.

The two commands:

both display a graph showing the 0.95-confidence ellipse for C(1) and C(2), C(1) and C(3), and C(2) and C(3).

displays multiple confidence ellipses (contours) for C(1) and C(2).

See “Confidence Intervals and Confidence Ellipses” for discussion.

See also Pool::wald.

Removes the pool’s history attribute, as shown in the label view of the pool.

The first line removes the history from the pool P1, and the second line displays the label view of P1, including the now blank history field.

See “Labeling Objects” for a discussion of labels and display names.

See also Pool::label.

Removes the pool’s remarks attribute, as shown in the label view of the pool.

The first line removes the remarks from the pool P1, and the second line displays the label view of P1, including the now blank remarks field.

See “Labeling Objects” for a discussion of labels and display names.

See also Pool::label.

Displays the covariances of the coefficient estimates for an estimated pool object.

displays the coefficient covariance matrix for POOL1 in a window. To store the coefficient covariance matrix as a sym object, use “@coefcov”:

See also Coef::coef.

Follow the pool name with the coint keyword, any options, and a list of two or more ordinary or pool series.

You may specify the type using one of the following keywords:

Depending on the type selected above, the following may be used to indicate deterministic trends:

Additional options:

performs a Johansen test for pool series Y?, X1?, and X2? with a lag of 1 to 2 and linear trend in the data, and an intercept but no trend in the cointegrating equation is assumed as exogenous variables.

See “References” for details on panel cointegration testing. See also Pool::uroot.

Creates either a named or unnamed copy of the pool.

creates an unnamed copy of the pool P1.

creates P2, a copy of the pool P1.

See also copy.

List the cross section identifiers after the define keyword.

The first line declares a pool object named SPOT with cross section identifiers UK, JPN, GER, and CAN. The second line redefines the identifiers to be UK, GER, ITA, and FRA.

See “Pooled Time Series, Cross-Section Data” for a discussion of cross-section identifiers.

See also Pool::add, Pool::drop and Pool::pool.

Follow the keyword by a list of the names of any series you wish to remove from the current workfile. Deleting does not remove objects that have been stored on disk in EViews database files.

The delete command allows you to delete series from the workfile using ordinary and pool series names.

You can delete an object from a database by prefixing the name with the database name and a double colon. You can use a pattern to delete all objects from a workfile or database with names that match the pattern. Use the “?” to match any one character and the “*” to match zero or more characters.

If you use delete in a program file, EViews will delete the listed objects without prompting you to confirm each deletion.

To delete all series in the workfile with names beginning with “CPI” that are followed by identifiers in the pool object MYPOOL:

See “Object Basics” for a discussion of working with objects, and “EViews Databases” for a discussion of EViews databases.

List the name of ordinary and pool series for which you wish to compute descriptive statistics.

By default, statistics are computed for each stacked pool series, using only common observations where all of the cross-sections for a given series have nonmissing data. A missing observation for a series in any one cross-section causes that observation to be dropped for all cross-sections for the corresponding series. You may change the default treatment of NAs using the “i” and “b” options.

EViews also allows you to compute statistics with the cross-section means removed, statistics for each cross-sectional series in a pool series, and statistics for each period, taken across all cross-section units.

displays the “within” descriptive statistics of the three series GDP, INV, CPI for the POOL1 cross-section members.

computes the statistics for GDP for each period, taken across each of the cross-section identifiers.

See “Pooled Time Series, Cross-Section Data” for a discussion of the computation of these statistics, and a description of individual and balanced samples.

Display the contents of a table, graph, or spool in the window of the pool object.

Display the contents of the table TAB1 in the window of the object POOL1.

Most often used in constructing an EViews Add-in. See “Custom Object Output”.

Attaches a display name to a pool object which may be used to label output in place of the standard pool object name.

Display names are case-sensitive, and may contain a variety of characters, such as spaces, that are not allowed in pool object names.

The first line attaches a display name “Hours Worked” to the pool object HRS, and the second line displays the label view of HRS, including its display name.

See “Labeling Objects” for a discussion of labels and display names.

See also Pool::label.

List the cross-section members to be dropped from the pool.

drops the cross-section members JPN, KOR, and HK from the pool CROSSSC.

“Cross-section Identifiers” discusses pool identifiers.

See also Pool::add.

fetch reads one or more objects from EViews databases or databank files into the active workfile. The objects are loaded into the workfile using the object in the database or using the databank file name. EViews will first expand the list of series using the pool operator, and then perform the fetch.

If you fetch a series into a workfile with a different frequency, EViews will automatically apply the frequency conversion method attached to the series by setconvert. If the series does not have an attached conversion method, EViews will use the method set by Options/Date-Frequency in the main menu. You can override the conversion method by specifying an explicit conversion method option.

The fetch command keyword is followed by a list of object names separated by spaces. The default behavior is to fetch the objects from the default database (this is a change from versions of EViews prior to EViews 3.x where the default was to fetch from individual databank files).

You can precede the object name with a database name and the double colon “::” to indicate a specific database source. If you specify the database name as an option in parentheses (see below), all objects without an explicit database prefix will be fetched from the specified database. You may optionally fetch from individual databank files or search among registered databases.

You may use wild card characters, “?” (to match a single character) or “*” (to match zero or more characters), in the object name list. All objects with names matching the pattern will be fetched.

To fetch from individual databank files that are not in the default path, you should include an explicit path. If you have more than one object with the same file name (for example, an equation and a series named CONS), then you should supply the full object file name including identifying extensions.

The database specified by the double colon “::” takes precedence over the database specified by the “d=” option.

In addition, there are a number of options for controlling automatic frequency conversion when performing a fetch. The following options control the frequency conversion method when copying series and group objects to a workfile, converting from low to high frequency:

The following options control the frequency conversion method when copying series and group objects to a workfile, converting from high to low frequency:

If no conversion method is specified, the series-specific or global default conversion method will be employed.

To fetch M1, GDP, and UNEMP pool series from the default database, use:

To fetch M1 and GDP from the US1 database and UNEMP from the MACRO database, use the command:

Use the “notifyillegal” option to display a dialog when fetching the series MYILLEG@LNAME that will suggest a valid name and give you to opportunity to name the object before it is inserted into a workfile:

See “EViews Databases” for a discussion of databases, databank files, and frequency conversion. Appendix A. “Wildcards” describes the use of wildcard characters.

See also Series::setconvert, Pool::store, and Pool::store.

Only valid for panel or pool regression equations estimated with fixed effects. Not currently available for specifications estimated using instrumental variables.

tests whether the fixed effects are jointly significant.

See “Fixed Effects Testing” for discussion. See also Pool::testadd, Pool::testdrop, Pool::ranhaus, and Pool::wald.

This procedure allows you to generate multiple series using the cross-section identifiers in a pool.

You may use the cross section identifier “?” in the series name and/or in the expression on the right-hand side.

The commands,

are equivalent to generating separate series for each cross-section:

Similarly:

generates three series Y_US, Y_UK, Y_CAN that are the log differences from X_US. Note that Y_US=0.

It is worth noting that the pool genr command simply loops across the cross-section identifiers, performing the evaluations using the appropriate substitution. Thus, the command,

is equivalent to entering:

so that upon completion, the ordinary series Z will contain Y_CAN.

See “Pooled Time Series, Cross-Section Data” for a discussion of the computation of pools, and a description of individual and balanced samples.

See Series::series for a discussion of the expressions allowed in genr.

As a procedure, label changes the fields in the pool object label.

The first version of the command displays the label view of the pool object. The second version may be used to modify the label. Specify one of the following options along with optional text. If there is no text provided, the specified field will be cleared.

The following lines replace the remarks field of POOL1 with “Data from CPS 1988 March File”:

To append additional remarks to POOL1, and then to print the label view:

To clear and then set the units field, use:

See “Labeling Objects” for a discussion of labels.

See also Pool::displayname.

ls estimates cross-section weighed least squares, feasible GLS, and fixed and random effects models.

ls carries out pooled data estimation. Type the name of the dependent variable followed by one or more lists of regressors. The first list should contain ordinary and pool series that are restricted to have the same coefficient across all members of the pool. The second list, if provided, should contain pool variables that have different coefficients for each cross-section member of the pool. If there is a cross-section specific regressor list, the two lists must be separated by “@CXREG”. The third list, if provided, should contain pool variables that have different coefficients for each period. The list should be separated from the previous lists by “@PERREG”.

You may include AR terms as regressors in either the common or cross-section specific lists. AR terms are, however, not allowed for some estimation methods. MA terms are not supported.

estimates pooled OLS of DY? on a constant, INV?, EDU? and YEAR.

estimates a fixed effects model without restricting any of the coefficients to be the same across pool members.

“Basic Regression Analysis” and “Additional Regression Tools” discuss the various regression methods in greater depth.

See “Pooled Time Series, Cross-Section Data” for a discussion of pool estimation, and “Panel Estimation” for a discussion of panel equation estimation.

See “Special Expression Reference” for special terms that may be used in ls specifications.

See also Pool::tsls for instrumental variables estimation.

List the ordinary and pool series to be placed in the group. If specified, group_name should be the first option.

places the ordinary series Z, and all of the series represented by the pool series X? and Y?, in the group G1.

See “Making a Group of Pool Series” for details.

If you provide a name for the model in parentheses after the keyword, EViews will create the named model in the workfile. If you do not provide a name, EViews will open an untitled model window if the command is executed from the command line.

estimates a VAR and makes a model named POOLMOD from the estimated pool object. POOLMOD includes an assignment statement “ASSIGN @PREFIX S_”. Use the command “show poolmod” or “poolmod.spec” to open the POOLMOD window.

See “Models” for a discussion of specifying and solving models in EViews.

See also Model::merge and Model::solve.

Creates and saves residuals in the workfile from a pool object.

Follow the object name with a period and the makeresids keyword, then provide a list of names to be given to the stored residuals. You may use a cross section identifier “?” to specify a set of names.

The residuals of each pool member will have a name starting with “RES1_” and the cross-section identifier substituted for the “?”.

See “Residuals”.

You should provide options, a list of series names, an “@” separator, and a list of command names for the statistics you wish to compute. The series will have a name with the cross-section identifier “?” replaced by the statistic command.

computes the mean and standard deviation of the GDP_? and EDU_? series in each period (across the cross-section members) using the default common sample. The mean and standard deviation series will be named GDP_MEAN, EDU_MEAN, GDP_SD, and EDU_SD.

Computes the maximum and minimum values of the GDP_? series in each period using the balanced sample. The max and min series will be named GDP_MAX and GDP_MIN.

See “Pooled Time Series, Cross-Section Data” for details on the computation of these statistics and a discussion of the use of individual, common, and balanced samples in pool.

See also Pool::describe.

Creates a system out of the pool equation specification. Each cross-section in the pool will be used to form an equation. The pool variable y is the dependent variable. The [x1 x2 x3 ...] list consists of regressors with common coefficients in the system. The @cxreg list are regressors with different coefficients in each cross-section. The list of variables that follow @inst are the common instruments. The list of variables that follow @cxinst are the equation specific instruments.

Note that period specific coefficients and effects are not available in this routine.

creates a system named SYS1 with INV? as the dependent variable and a common intercept for each cross-section member. The regressor CAP? is restricted to have the same coefficient in each equation, while the VAL? regressor has a different coefficient in each equation.

This command creates a system named SYS2 with INV? as the dependent variable and a different intercept for each cross-section member equation. The regressor CAP? enters each equation with a different coefficient and each equation has two instrumental variables @TREND and INV? lagged.

See “System Estimation” for a discussion of system objects in EViews.

See “Object Linking and Embedding (OLE)” for a discussion of using OLE with EViews.

output changes the default object view to display the estimation output (equivalent to using Pool::results).

The output keyword may be used to change the default view of an estimation object. Entering the command:

displays the estimation output for pool POOL1.

See Pool::results.

Follow the pool keyword with a name for the pool object. You may optionally provide the identifiers for the cross-section members of the pool object. Pool identifiers may be added or removed at any time using Pool::add and Pool::drop.

Declares a pool object named ZOO1 with the listed cross-section identifiers.

See “Pooled Time Series, Cross-Section Data” for a discussion of working with pools in EViews.

See Pool::add and Pool::drop. See also Pool::ls for details on estimation using a pool object.

Tests the hypothesis that the random effects (components) are correlated with the right-hand side variables in a pool equation setting. Uses Hausman test methodology to compare the results from the estimated random effects specification and a corresponding fixed effects specification. If the estimated specification involves two-way random effects, three separate tests will be performed; one for each set of effects, and one for the joint effects.

Only valid for pool regression equations estimated with random effects. Note that the test results may be suspect in cases where robust standard errors are employed.

estimates a specification with cross-section random effects and tests whether the random effects are correlated with the right-hand side variables ADVER and LSALES using the Hausman test methodology.

See also Pool::testadd, Pool::testdrop, Pool::fixedtest, and Pool::wald.

May be used to import data into an existing workfile from a text, Excel, or Lotus file on disk.

Note: we strongly recommend that you instead of using this proc, you use wfopen or pageload to read the source data into a panel structured workfile and pageunstack if desired.

You must supply the name of the source file. If you do not include the optional path specification, EViews will look for the file in the default directory. Path specifications may point to local or network drives. If the path specification contains a space, you may enclose the entire expression in double quotation marks.

Follow the source file name with a list of ordinary or pool series.

If you do not specify the “t” option, EViews uses the file name extension to determine the file type. If you specify the “t” option, the file name extension will not be used to determine the file type.

reads stacked data from an ASCII file MYDAT.RAW in the A: drive. The data in the file are stacked by cross-section, the missing value NA is coded as a “.” (dot or period). We read one ordinary series YEAR, and three two pool series LWAGE? and HRS?.

reads data from an Excel file STATEPAN in the default directory. The data are stacked by period in the sheet SHEET3 with the upper left data cell A2. We read three pool series INC? EDUC? and POP?.

See “Creating a Workfile by Reading from a Foreign Data Source” and “Importing Data” for a discussion and examples of importing data from external files.

“Working with Panel Data” describes panel data alternatives to working with pooled data.

See also pageload and wfopen and Pool::write.

displays the specifications of the estimation object POOL1.

See “Estimating a Pool Equation” for a discussion of pool equations.

Displays the correlations of the residuals from each pool cross-section equation.

displays the residual correlation matrix of POOL1.

See also Pool::residcov and Pool::makeresids.

Displays the covariances of the residuals from each pool cross-section equation.

displays the residual covariance matrix of POOL1.

See “Estimating a Pool Equation” for a discussion of pool equations.

See also Pool::residcor and Pool::makeresids.

Display the actual, fitted values and residuals in either tabular or graphical form. resids displays multiple graphs, where each graph will contain the residuals for each cross-section in the pool.

regresses M1 on a constant, INC, and TB3, and displays a table of actual, fitted, and residual series.

displays a graph of the actual, fitted, and residual series.

See also Pool::makeresids.

See “Estimating a Pool Equation” for a discussion of pool equations.

estimates an equation using least squares, and displays and prints the results.

See “Estimating a Pool Equation” for a discussion of pool equations.

Sets the attribute attr to attr_value. Note that quoting the arguments may be required. Once added to an object, the attribute may be extracted using the @attr data member.

sets the “revised” attribute in the object A to the string “never”, and extracts the attribute into the string object S.

See “Adding Custom Attributes in the Label View” and “Adding Your Own Label Attributes”.

The sheet view displays the spreadsheet view of the series in the pool. Follow the word sheet by a list of series to display; you may use the cross section identifier “?” in the series name.

displays and prints the pool spreadsheet view of the series X?, Y?, and Z?.

See “Pooled Time Series, Cross-Section Data” for a discussion of pools.

Stores one or more objects in the current workfile in EViews databases or individual databank files on disk. The objects are stored under the name that appears in the workfile. EViews will first expand the list of series using the pool operator, and then perform the operation.

Follow the store command keyword with a list of object names (each separated by a space) that you wish to store. The default is to store the objects in the default database. (This behavior is a change from EViews 2 and earlier where the default was to store objects in individual databank files).

You may precede the object name with a database name and the double colon “::” to indicate a specific database. You can also specify the database name as an option in parentheses, in which case all objects without an explicit database name will be stored in the specified database.

You may use the wild card character “*” to match zero or more characters in the object name list. All objects with names matching the pattern will be stored. You may not use “?” as a wildcard character, since this conflicts with the pool identifier.

You can optionally choose to store the listed objects in individual databank files. To store in files other than the default path, you should include a path designation before the object name.

If you do not specify the precision option (1 or 2), the global option setting will be used. See “Database Storage Defaults”.

stores the three pool objects M1, GDP, UNEMP in the default database.

“Basic Data Handling” discusses exporting data in other file formats. See “EViews Databases” for a discussion of EViews databases and databank files.

For additional discussion of wildcards, see Appendix A. “Wildcards”.

See also Pool::fetch.

Tests the hypothesis that the listed variables were incorrectly omitted from an estimated equation (only available for equations estimated by list). The test displays some combination of Wald and LR test statistics, as well as the auxiliary regression.

List the names of the series to test for omission after the keyword.

tests the addition of the pool series GDP? to the common coefficients list and INC? to the cross-section specific coefficients list.

See “Coefficient Diagnostics” for further discussion.

See also Pool::testdrop and Pool::wald.

List the names of the series to test for omission after the keyword.

drops X? from the existing pool specification and prints the results of the test.

See “Coefficient Diagnostics” for further discussion of testing coefficients.

See also Pool::testadd and Pool::wald.

Type the name of the dependent variable followed by one or more lists of regressors. The first list should contain ordinary and pool series that are restricted to have the same coefficient across all members of the pool. The second list, if provided, should contain pool variables that have different coefficients for each cross-section member of the pool. If there is a cross-section specific regressor list, the two lists must be separated by “@CXREG”. The third list, if provided, should contain pool variables that have different coefficients for each period. The list should be separated from the previous lists by “@PERREG”.

You may include AR terms as regressors in either the common or cross-section specific lists. AR terms are, however, not allowed for some estimation methods. MA terms are not supported.

Instruments should be specified in one of three lists. The “@INST” list should contain instruments that are common across all cross-sections and periods. The “@CXINST” should contain instruments that differ across cross-sections, while the “@PERINST” list specifies instruments that differ across periods.

There must be at least as many instrumental variables as there are independent variables. All exogenous variables included in the regressor list should also be included in the corresponding instrument list. A constant is included in the common instrumental list if not explicitly specified.

estimates TSLS on the pool specification using common instruments Z?

See “Two-stage Least Squares” and “Two-Stage Least Squares” for details on two-stage least squares estimation in single equations and systems, respectively.

“Instrumental Variables” discusses estimation using pool objects, while “Instrumental Variables Estimation” discusses estimation in panel structured workfiles.

See also Pool::ls.

Copies coefficients from the pool into the appropriate coefficient vector.

Follow the name of the pool object by a period and the keyword updatecoef.

places the coefficients from pool1 in the default coefficient vector C.

See also Coef::coef.

When used with a pool series, the procedure will perform panel unit root testing. The panel unit root tests include Levin, Lin and Chu (LLC), Breitung, Im, Pesaran, and Shin (IPS), Fisher - ADF, Fisher - PP, and Hadri tests on levels, or first or second differences.

Enter the pool object name followed by a period, the keyword, and the name of a pool “?” series.

You should specify the exogenous variables and order of dependent variable differencing in the test equation using the following options:

You may use one of the following keywords to specify the test:

Specifies the number of lag difference terms to be included in the test equation. Applicable in “Summary”, LLC, Breitung, IPS, and Fisher-ADF tests. The default setting depends on whether you choose to balance the samples across cross-sections.

If you do not include the “balance” option, the default is to perform automatic lag selection using the Schwarz criteria (“lagmethod=sic”).

Alternately, if you include the “balance” option, the default setting is a common, observation-based fixed lag (“lag=default”) where:

Specifies options for computing kernel estimates of the zero-frequency spectrum (long-run covariance). Applicable to “Summary”, LLC, Fisher-PP, and Hadri tests.

performs the LLC panel unit root test with exogenous individual trends and individual effects on pool series GDP?

performs the IPS panel unit root test on pool series INV?. The test includes individual effects, lag will be chosen by AIC from maximum lag of three.

performs a summary of the panel unit root tests on the pool series MM?. The test equation includes a constant term and three lagged first-difference terms. The frequency zero spectrum is estimated using kernel methods (with a Parzen kernel), and a bandwidth of 2.3.

See “Unit Root Testing” for discussion of standard unit root tests performed on a single series, and “Cross-sectionally Independent Panel Unit Root Testing” and “Cross-sectionally Dependent Panel Unit Root Tests” for discussion of unit roots tests performed on panel structured workfiles, groups of series, or pooled data.

See also Pool::uroot2, Series::uroot, Series::uroot2, Group::uroot, Group::uroot2.

where group_name is the name of a pool object and pool_series is a pool series.

The line above performs a PANIC unit root test on the series in the pool series SER?.

The line above performs a CIPS unit root test on the pool series SER?, with ADF testing performed on each cross-section with a constant and trend, and ADF lag selection using the Schwarz criterion.

See “Unit Root Testing” for discussion of standard unit root tests performed on a single series, and “Cross-sectionally Independent Panel Unit Root Testing” and “Cross-sectionally Dependent Panel Unit Root Tests” for discussion of unit roots tests performed on panel structured workfiles, groups of series, or pooled data.

See also Pool::uroot, Series::uroot, Series::uroot2, Group::uroot, Group::uroot2.

The wald view carries out a Wald test of coefficient restrictions for a pool object.

Enter the pool object name, followed by a period, and the keyword. You must provide a list of the coefficient restrictions, with joint (multiple) coefficient restrictions separated by commas.

declares a pool object with three cross section members (US, UK, JPN), estimates a pooled OLS regression with separate AR(1) coefficients, and tests the null hypothesis that all AR(1) coefficients are equal.

See “Wald Test (Coefficient Restrictions)” for a discussion of Wald tests.

See also Pool::cellipse, Pool::testdrop, Pool::testadd.

Creates a foreign format disk file containing EViews data. May be used to export EViews data to another program.

Note: we strongly recommend that you instead of using this proc, you use pagestack to create a panel structured workfile and then use wfsave or pagesave.

Follow the keyword by a name for the output file and list the series to be written. The optional path name may be on the local machine, or may point to a network drive. If the path name contains spaces, enclose the entire expression in double quotation marks.

Note that EViews cannot, at present, write into an existing file. The file that you select will, if it exists, be replaced.

Other options are used to specify the format of the output file.

If you omit the “t=” option, EViews will determine the type based on the file extension. Unrecognized extensions will be treated as ASCII files. For Lotus and Excel spreadsheet files specified without the “t=” option, EViews will automatically append the appropriate extension if it is not otherwise specified.

Writes into an ASCII file named “Dat1.csv” on the A drive. The data file is listed by observations, NAs are coded as “.” (dot), each series is separated by a comma, and the date/observation numbers and cross-section identifiers are written together with the series names.

writes the same file in the default directory.

writes an Excel file “Growth.xls” in the specified directory. The data are organized by observation, and are listed by period/time.

See “Exporting Data” for a discussion. Pool writing is discussed in “Exporting Pooled Data”.

See also pagesave and Pool::read.