WESVAR

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

This list of FAQs will be updated periodically.

Q:	Another analyst is using a workbook that I created. He is receiving a "file not found" error when opening the workbook. Why?
A:	If the workbook specifies the data file as, say, M:\mydata.var, and the other user has the same directory mapped as N: (or another letter), WesVar will not be able to find the data file. You should use UNC filenames, like \\SRV1\VOL23\Myprojects\mydata.var in workbooks if sharing workbook files on a network, in the Input Data fill-in box in the workbook.
Q:	Does the order of the factors in the external FPC and JKN factor files matter?
A:	Yes. The factor files have one record per replicate. The records must be in the same order as the physical order of the replicate weights on the input file used to create the data (VAR) file. For example, suppose your replicate weights are named RWT1-RWT100, but that they were sorted in alphanumeric order: RWT1, RWT10, RWT100, RWT11, etc. The factor files must be in this order also. In version 4.2, the attach factors screen has a new format. Replicate weights are listed on the Attach Factors screen by their physical order in the data file. Both FPC factors and JKn factors must appear in this same order on the external files.
Q:	I use a non-English keyboard and cannot enter a decimal value using a comma for Fay_K on the Data File screen or for the logistic convergence criterion in Preferences or in a Regression Request/Options screen. What can I do?
A:	This problem has been fixed in version 4.2.
Q:	What does the following error mean in logistic regression: "Estimates are converging slowly or are diverging for full sample and replicates..."?
A:	This message is issued in logistic regression when the convergence criterion for parameter estimates is not met in the specified maximum number of iterations. This could be caused by the following: The convergence criterion has been changed from difference in log-likelihood in version 3 to relative difference in log-likelihood in version 4. If you were a version 3 user, you may have to manually change the convergence criterion from 0.025 to 1E-6 in Preferences and in any version 3 workbook that you import into version 4. You have few (or no) cases that have the characteristic being modeled in a cell. If your model contains a categorical independent variable, say RACE, and in replicate subsample 43 for RACE.1 there are no cases with the characteristic (dependent variable = 1), then the estimation algorithm will try to send the parameter estimate to negative infinity so that the predicted probability for RACE.1 is 0. In that case the parameter estimate will not converge but will get more and more negative at each iteration. You may need to collapse categories for one of your independent variables to eliminate the problem. In the Regression Request/Output Control, check the boxes for Iteration History and Replicate Coefficients to get a detailed report on how the parameter estimates are changing.
Q:	I can't get WesVar to read the SAS transport files I created. How do I create the transport file so that it's acceptable to WesVar?
A:	One alternative is to use the following code to create your SAS transport files. This is code for creating a SAS transport file (with a single member) named c:\export\wesvar.xpt from the sd2 file c:\input\data.sd2 to import into WesVar 4.0:    options validvarname=v6; libname in1 'c:\input'; libname out1 xport 'c:\export\wesvar.xpt'; proc copy in=in1 out=out1; select data; run; The statement, options validvarname=v6;, will cause SAS to generate unique 8-character names for variables with names longer than 8 characters. The VALIDVARNAME option is only valid using SAS Version 7 or later. It is not recognized (or needed) by earlier versions of SAS.
Q:	I'm having trouble installing WesVar on Windows NT or Windows 2000. How can I get it to work?
A:	There does not seem to be a problem if the WesVar user has Administrator privileges.
Q:	I converted a WesVar 4.0 data file to SAS XPORT format. Using WesVar, I re-imported the data into WesVar and seemed to lose all my value labels. Did I do something wrong or can value labels not be imported?
A:	SAS value labels are not imported in WesVar, even if they were originally created in WesVar and exported to SAS.
Q:	What does the following error mean: "STRATUM 1 has non-consecutive varunits for METHOD=JKn"?
A:	VARUNITS must be numbered consecutively starting at 1 in each VARSTRAT. The number of VARUNITS can differ from one VARSTRAT to another.
Q:	Is it possible to create weights with one PSU in some strata with JKn design?
A:	WesVar does not allow the JKn method to be used unless there are at least 2 PSUs in every stratum. To use the JKn method, you would have to collapse any stratum with 1 PSU with another stratum.

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏0 回帖

关键词：WESVAR SVAR SVA VaR periodically WESVAR

http://www.westat.com/wesvar/techpapers/ACS-Replication.pdf

Analysis of Complex Sample Data

Using Replication

J. Michael Brick

Wesvar 4.3 User's Guide

http://www.westat.com/wesvar/about/WV_4-3_Manual.pdf

好东东

WesVar Version 4 is available free of charge.

Download Version 4 (free)

Download Version 4 (free)

WESVAR: SOFTWARE FOR COMPLEX SURVEY DATA ANALYSIS

Using bootstrap weights with Wes Var and SUDAAN
By Owen Phillips

http://www.statcan.ca/english/freepub/12-002-XIE/2004002/pdf/phillips.pdf

Steps to Define a Regression Model:

Step 1 Expand the Regression Request node on the workbook tree and click the Models node.  The Regression Model panel is displayed (see Figure 6-5).

Step 2 Specify the dependent and independent variables by moving them from the Class Variables or Source Variables list as described below.

 
Figure 6-5.  Regression Model Panel

Dependent.  For the dependent variable, you can select numeric variables from the Source Variables list only.  In linear regression, you can also specify a transformation (e.g., Exp, Log10, or Sqrt) of a source variable as the dependent variable.  If a variable has been used as the dependent variable, it cannot be used as an independent variable.  For logistic models, you can select a character variable from the Source Variables box for the dependent variable as long as the variable has only two categories.  For a multinomial logistic model, you can select a numeric or character variable as long as the number of levels or distinct values is 256 or fewer.  Class variables cannot be used as dependent variables.

Independent.  For the independent variable, you can select numeric variables from the Source Variables list or class variables from the Class Variables list.  Class variables provide a simplified method for specifying categorical variables as independent variables.  A variable can appear in the model either as a class variable or as a continuous variable (from the source list), but not both.  A variable is only allowed as an independent variable once, but there is no limit on the number of independent variables that you can specify.  WesVar will display a message if there are more regressors than degrees of freedom.  You can specify functions of independent variables by using the calculator pad and the Functions drop-down box below the variable boxes (see step 3).  Transformations of variables such as exp(x) can be used as independent variables in a model.  However, products such as Z*exp(x) cannot be used directly.  If such a product is desired, you should use the New Continuous [to Continuous] recode feature to create the product.

 Note Any observation with a missing value for any variable specified in the model is deleted from the analysis of that model.
Step 3 Variables can also be included as part of an interaction term.  To specify a model with interactions, double-click the first variable, click the multiplication button (? on the calculator pad, and then double-click the second variable.  Two-way and three-way interactions are allowed.  You can specify higher order interactions by creating the appropriate crossproduct terms by using Recode.  For more details on recodes, see Chapter 4.  For example, for a four-way interaction term, create a new variable that is equal to the product of a variable and a three-way interaction.  The four-way interaction is, thus, reduced to a two-way interaction using the newly created variable.

 Note When creating interactions, we recommend including all interactions and main effects in the model.  See Appendix C for more information on the parameterization and its effects on model interpretation.
Step 4 Specify Intercept and Df options.  To include the intercept term from a linear or logistic model, click the Intercept box.  You can specify a model with only the intercept term.  The model will appear on the tree as Dependent variable =.  The Df option lets you change the number of degrees of freedom associated with the variance-covariance matrix.  As the default, Df is equal to the number of replicates.  For more information on the degrees of freedom, see Appendix C.

Step 5 Select Stnd. Coef. for standardized regression coefficients.  Standardized coefficients will be produced as well as regular coefficients for linear regression models only.  Additionally, if the specified model contains any interaction terms, an error message will appear and standardized coefficients will not be estimated.  For more information about standardized regression coefficients, see Appendix燙.

Step 6 Click the Add as New Entry button.  The item appears as a new node under the Models node of the workbook tree.  If you want to change a previously defined model, click the model on the tree, change your dependent or independent variables, and click the Replace Current Entry button.  Hypothesis tests that no longer apply will be deleted and starting values for a logistic model will be set back to zero.

To delete or remove a variable from the model, click the model on the tree.  Then right-click the variable in the Independent or Dependent variable box and choose the Delete option from the popup menu.  More than one model can be specified for each request.  However, a request cannot include both a linear and a logistic model.

 Note Putting multiple models in the same request (rather than one model per request) will often speed calculations.

thansk