Showing papers on "Unit-weighted regression published in 1987"

Regression techniques for the detection of analytical bias

Abstract: Regression techniques are commonly applied to compare two analytical methods at several concentrations and to test the biases of one method relative to another. However, regression is strictly applicable only when one method is without error, for example in comparisons with reference materials. A regression-like technique, maximum-likelihood fitting of a functional relationship (MLFR), is explained and is demonstrated to work well. Under some conditions weighted regression provides a good approximation to MLFR, and so can be used if more convenient.

Error modeling and confidence interval estimation for inductively coupled plasma calibration curves

Abstract: : A simple linear calibration function can be used over a wide concentration range for the Inductively Coupled Plasma (ICP) spectrometer due to its linear responses. The random errors over wide concentration ranges are not constant, and constant variance regression should not be used to estimate the calibration function. Weighted regression techniques are appropriate if the proper weights can be obtained. Use of the calibration curve to estimate. The concentration of one or more unknown samples is straightforward, but confidence interval estimation for multiple use of the calibration curve is les obvious. A method is described for modeling the error along the ICP calibration curve using the estimated parameters from the fitted model to calculate weights for the calibration curve fit. Multiple and single-use confidence interval estimates are obtained and results along the calibration curve are compared.

Regression Analysis of State Role in Press Economics

Abstract: Conclusion These results indicate a slow and general shift by the nation’s leading periodicals to present environmental news in perhaps more meaningful ways. An increased complexity of coverage in some of the nation’s leading periodicals, and a general inclusion of more economic/ developmental coverage is evidence of this. The New York Times’ coverage of environmental reporting seems to be the only major exception over the 12-year period, instead of juxtaposing aspects of environmental issues, the Times apparently separated them. As the “national newspaper of the elite,”l5 this exception may have serious implications. Another shift across time to relatively fewer articles on degradation and protection of the natural environment, and to more on economic and developmental issues may reflect a broad movement toward environmental deregulation and economic development in more recent years. That the story bias has during the same period shifted from environmental to neutral may be indicative of a more rational, balanced or better informed approach toward environmental issues.

Applied GEWR(n,p,q) Normal Discount Bayesian Model: An Austrian Economic Case Study

Abstract: The theory of Generalised Exponentially Weighted Regression (GEWR) and dynamic Bayesian models has been given previously by Harrison-Akram(1982), Akram-Harrison(1983) and Akram(1984). This paper breifly reviews some of the main results and applies them to seasonal data concerned with the disposable personal income in Austria. For the selection of an appropriate model a new Stepwise Identification Procedure(SIP) based on a nonparametric measure, called Average String Length(ASL), is used. Both short and long term full forecasts and trends are obtained from a single model using on-line Bayesian learning procedure. The model applied yields optimum forecasts in the senses of minimum mean square error and whiteness of one step ahead forecast errors.

Weighted least square estimates of the parameters of a model of survivorship probabilities.

Abstract: A weighted regression has been fitted to estimate the parameters of a model involving functions of survivorship probability and age. Earlier the parameters were estimated by the method of ordinary least squares and the results were very encouraging. However a multiple regression equation passing through the origin has been found appropriate for the present model from statistical consideration. Fortunately this method while methodologically more sophisticated has a slight edge over the former as evidenced by the respective measures of reproducibility in the model and actual life tables selected for this study. (EXCERPT)

A Weighted Regression Model for Ratemaking

Abstract: This article develops and demonstrates the use of a weighted regression model for ratemaking. This model is superior to the traditional one which is based on the central limit theorem because it holds neither the mean nor the variance of the pure premium distribution constant over time. Comparison of actual premiums versus projected ones shows that the model is capable of producing adequate and fairly reasonable rate forecasts. By using the model the development and revision of rates can be perceived as an objective process.