| Main Title |
Alternative to Least Squares Statistics for Comparison of Duplicate Measurements. |
| Author |
Smith, Marjolein V. ;
Shaw, Jr., Robert W. ;
Paur, Richard J. ;
|
| CORP Author |
Environmental Sciences Research Lab., Research Triangle Park, NC. |
| Year Published |
1983 |
| Report Number |
EPA-600/J-81-083; |
| Stock Number |
PB83-241216 |
| Additional Subjects |
Least square method ;
Measuring instruments ;
Linear regression ;
Comparison ;
Chemical analysis ;
Substitutes ;
Environmental surveys ;
Reprints ;
Bartlett method ;
Numerical solution
|
| Holdings |
| Library |
Call Number |
Additional Info |
Location |
Last
Modified |
Checkout
Status |
| NTIS |
PB83-241216 |
Some EPA libraries have a fiche copy filed under the call number shown. |
|
07/26/2022 |
|
| Collation |
10p |
| Abstract |
The authors describe a method for the comparison of measurements of the same pollutant taken simultaneously by two instruments. The method requres no information about the precision of the instruments and does not assume, as the typically used linear regression by least squares does, that the errors for one instrument are negligible. The method does require that the data be distributed to permit unambiguous separation into three groups. The method provides confidence interval estimates. They present examples and recommendations for using the method and compare it with least squares. The body of information presented in this paper is directed to anyone who uses linear regression. |
