Finear model

The ANOVA results for these models are given in Table 6A8. Neither the linear nor the quadratic model shows significant lack of fit and both are significant at the 95% confidence level. The finear model should be preferred, because it is simpler. 

SHELFLIFE.dat The content (% of nominal) of two active components in a dosage form was assayed at various times (0-60 months) during a pharmaceutical stability trial to determine the acceptable shelf-life of the formulation the point at which the lower 90% confidence limit of the finear regression model intersects the 90%-of-nominal line gives the answer. Use with SHELFLIFE or LINREG. 

For the simplest case of univariate input variables and univariate output measures the following model describes a finear dependency between the input (or control) variable X = and the output variable y = (j/i,...,j/r)"... 

For breast and thyroid cancer as well as for leukemia, different models were chosen by BEIR VII. The breast cancer model is based on a pooled analysis of data from eight breast cancer incidence studies by Preston et al. (2002). The thyroid cancer model is based on a pooled analysis of data from seven thyroid cancer incidence studies by Ron et al. (1995). For leukemia, the BEIR VII model includes a time, t (= a-e), since exposure dependence (decreasing with increasing t) and a finear-qua-dratic function of dose. 

---