SOLICITATION NOTICE
B -- SPECIAL STUDIES AND ANALYSIS, LINEAR UNBIASED PREDICTION FOR SMALL DOMAIN
- Notice Date
- 7/14/2004
- Notice Type
- Solicitation Notice
- NAICS
- 541990
— All Other Professional, Scientific, and Technical Services
- Contracting Office
- Department of Health and Human Services, Center for Disease Control and Prevention, Acquisition and Assistance Field Branch (Pittsburgh), Post Office Box 18070 Cochrans Mill Road, Pittsburgh, PA, 15236-0070
- ZIP Code
- 15236-0070
- Solicitation Number
- MLM000HCS13-2004-10275
- Response Due
- 7/28/2004
- Archive Date
- 8/12/2004
- Point of Contact
- Margaret Mooney, Purchasing Agent, Phone (412)386-6431, Fax (412)386-6843, - John Columbia, Contract Specialist, Phone 412-386-4458, Fax 412-386-6429,
- E-Mail Address
-
zia3@cdc.gov, akq8@cdc.gov
- Description
- The Centers for Disease Control and Prevention (CDC) intend to negotiate on a sole source basis with the University of Rochester, Dr. S.R.S. Rao Produri to perform a professional service contract for Linear Unbiased Prediction For Small Domains. The PSC is a continuation of previous research on small domain estimators; therefore, this announcement is being issued for informational purposes only. This acquisition is being processed under FAR Part 13, Simplified Acquisition Procedures BACKGROUND INFORMATION Several approaches for small domain estimation employing data collected in large and small scale surveys have been proposed and described in statistical literature. A substantial body of research on Bayesian approaches, including empirical Bayes and hierarchical Bayesian models, now exists. The research to be conducted under this PSC differs from other ongoing research in that it involves extensions of classical random-effects models and the attending estimation theory with a special focus on evaluating the estimators for individual small domains. Thus, the research emphasis is on how well a small domain estimator performs in each instance where an estimate is needed and not on how well an estimator performs on average overall. This PSC is a continuation of previous research on small domain estimators in which the following features of Best Linear Unbiased Predictors (BLUP’s) and Empirical BLUP’s (or EBLUP’s) were investigated: 1. Unconditional biases and mean-square-errors (MSE's) of the BLUPs and EBLUPs and their expected values. Comparison of these MSEs with the variance of the sample mean. 2. Expected values of the conditional biases and MSE's of the BLUPs and EBLUPs. 3. Conditional and unconditional biases and MSEs of the BLUPs and EBLUPs for estimating the mean of a small domain. Both the cases of equal and unequal sample sizes for small domains were considered. It is essential to note that some of the results on the biases and MSEs were obtained through ad hoc approximations which require further study. OBJECTIVES As discussed above, in the completed research, biases and MSEs of the BLUPs and EBLUPs were examined through approximations to their expressions. Further examination of the biases and MSEs will be conducted under this PSC. The biases and MSEs of the BLUPs and EBLUPs for the small domain means will be examined for unequal sample sizes and some consideration will be given to the randomness of the number of sample units for a small domain. This is important since the sample sizes for the small domains are usually random and unequal. The unconditional MSE of the BLUP and the average of the unconditional MSEs of the EBLUPs have been examined in the literature. The conditional and unconditional biases and MSEs of the EBLUPs will be examined. This topic is of central importance since the mean of each small domain must be estimated. Estimation of the small domain means can be improved through the regression-type of models. Regression models for BLUP and EBLUP estimators will be proposed and studied. This part of the research will bring out the additional benefits of using auxiliary information for the BLUPs and EBLUPs of small domain means. SCOPE OF WORK To achieve the above objectives, the following tasks will be carried out in collaboration with the Project Officer: (1) The approximations to the biases and MSEs of the BLUPs and EBLUPs will be evaluated by considering additional terms in the respective expressions. (2) The conditional and unconditional biases and MSEs of the BLUPs and EBLUPs for the individual small domain means for the case of unequal sample sizes will be examined. The biases and MSEs of the estimators and the ratios of the squared biases to the MSEs will be examined for different patterns of the sample sizes. (3) Regression type of estimators and the corresponding BLUPs and EBLUPs for small domains will be suggested, and their properties will be evaluated. The average biases and MSEs of the estimators will be examined. Further, biases and MSEs of the estimators for the individual means will be evaluated To be considered qualified, sources must submit a capabilities statement which demonstrates in writing their ability to meet the following requirements: 1. Provide detailed and extensive knowledge of linear model theory covering conditional and unconditional biases and MSE’s of BLUPs and EBLUPs and mixed model estimators. 2. Formulate and solve problems in small domain estimation arising in the context of data analysis for public health surveys. 3. Develop design-based, model-based and model-assisted estimation strategies for use with data from national sample surveys. All of the above requirements should be supported by a record of completed and funded research projects. Evidence of the required formal theoretical justification for that work can be provided by citation of books, articles in learned journals and research reports on the subjects of interest. Qualified organizations are encouraged to submit a capabilities statement which addresses the requirements and contains material in sufficient detail to allow the CDC to determine if the party can perform this requirement. Capabilities are to be received in the contracting office no later than fifteen (15) days from the date of this announcement. Submit written information to: Margaret Mooney, MS-P05, Reference: MLM000HCS13-2004-10275, CDC/PGO/AAFB, 626 Cochrans Mill Road, PO Box 18070, Pittsburgh, PA 15236-0070 or responses may be submitted electronically to Margaret Mooney at zia3@cdc.gov The intent of this synopsis is to determine whether alternative sources exist. Information received will be used solely for the purpose of determining whether to conduct a competitive procurement. A determination by the Government not to compete this proposed requirement based upon responses to this notice is solely within the discretion of the Government. All responsible sources may submit a response, which shall be considered by the Agency. If any responses are submitted in response to this presolicitation notice and the agency determines to compete this requirement, an RFQ will be issued solely to the responders of this notice. A subsequent presolicitation notice will not be issued. NOTE: THIS NOTICE WAS NOT POSTED TO WWW.FEDBIZOPPS.GOV ON THE DATE INDICATED IN THE NOTICE ITSELF (14-JUL-2004); HOWEVER, IT DID APPEAR IN THE FEDBIZOPPS FTP FEED ON THIS DATE. PLEASE CONTACT fbo.support@gsa.gov REGARDING THIS ISSUE.
- Web Link
-
Link to FedBizOpps document.
(http://www.eps.gov/spg/HHS/CDCP/CMBP/MLM000HCS13-2004-10275/listing.html)
- Record
- SN00620864-F 20040716/040714213250 (fbodaily.com)
- Source
-
FedBizOpps.gov Link to This Notice
(may not be valid after Archive Date)
| FSG Index | This Issue's Index | Today's FBO Daily Index Page |