Regression mean response

A valuable inference that can be made to infer the quality of the model predictions is the (l-a)I00% confidence interval of the predicted mean response at x0. It should be noted that the predicted mean response of the linear regression model at x0 is y0 = F(x0)k or simply y0 = X0k. Although the error term e0 is not included, there is some uncertainty in the predicted mean response due to the uncertainty in k. Under the usual assumptions of normality and independence, the covariance matrix of the predicted mean response is given by... 

Ensure that the responses from samples are close to the mean response, y, of the calibration set. This will decrease the error contribution from the least-squares estimate of the regression line. 

Although calibration and regression are closely related, they do not coincide with each other. With regression, the response Y, is estimated from the independent variable Xt by means of the mathematical model Yt = f(X,) +On the other hand, in analytical chemistry the independent variable (that is, the concentration) is predicted from the response variable (i.e., the instmment response). One should be aware that, from a statistical viewpoint, this is not correct. 

The idea of the lack-of-fit test is to compare the pure error of the regression line with the error due to the use of an inappropriate regression model. The MSLOf, which is a measure for the spread of the mean response per concentration from the regression line, is divided by the MSpE, which is a measure for the spread of the instrument response due to experimental variation. The obtained F-value (F = MS[ oii/MS i) is compared with the F-distribution with k 2 and n k d.f. 

The goal of regression analysis is usually two-fold. First, a model is needed to explain the data. Second, using the model, predictions about mean responses or future observations may be needed. The distinction between mean responses and future observations must be clarified. Mean responses are based on already observed... 

The NIPALS algorithm extracts one factor (a principal component) at a time from the mean-centred data matrix. Each factor is obtained iteratively by repeated regression of response (absorbance) data, F , on the scores (principal components) Z to obtain improved loadings (eigenvectors) V, and of F on V to obtain improved Z. 

The regression mean square MSgECR i the sum of squares per coefficient, explained by the model. It is associated, as is the corresponding sum of squares, with p - 1 degrees of freedom. If the dependence of the response on the model parameters were simply a result of chance fluctuations the regression mean square would also be an estimate of the variance. In this case regression and residual means squares would be expected to be similar, and their ratio to be around one. 

At times, the researcher wants to predict y based on specific x, values. In estimating a mean response for Y, one needs to specify a vector of x, values within the range in which the y model was constructed. For example, in Example 4.2, looking at the regression equation that resulted when the x, were added to the model (forward selection), we finished with... 

There are two types of prediction intervals which can be constructed in the regression problem, prediction intervals for a population mean (the mean response y for a given x ) and prediction intervals for individual observations (i.e. the prediction interval for a particular patient). Conceptually, the difference between the two intervals is subtle. In the first case, one is interested in an interval for the population mean at a given value of X. In the second case, an individual observation from the population is of interest. In practice the difference between the two can be substantial since the prediction interval for the population mean is more narrow than that for an individual. To illustrate this difference, consider... 

Laboratory and/or field data were analyzed using SAS systems (SAS Institute Inc. 1999) utilizing analysis of variance, regression analysis, response surface analysis, univariate analysis, repeated measures analysis (multivariate profile analysis), covariance analysis and/or principle components analysis. Good statistical practices were used to verify that the data satisfied the assumptions underlying the various analyses. Significant differences between means were determined by Tukey s Studentized Range Test, the Tukey-Kramer HSD test, or the Bonferroni t test. Alpha was set at 0.05. 

Proof This requires showing two things the mean response value and its associated confidence interval. First, consider the mean response value given by E(y xd) = PY xd - The best estimate for this value is the mean value obtained from the regression equation, that is. 

The general regression modeling approach takes the standard perspective in much of statistical modeling of focusing directly on the mean responses and how they change over time (Phoebus 1986). 

CA 125 is a widely used cancer marker for monitoring treatment responses and detecting disease recurrences in patients with ovarian cancer. Generally, the CA 125 cutoff value of 35 U/ml is used for the mean value in normal women. It has been shown that values above and below this cutoff correlate reasonably well with the regression or progression of disease (6, 7). 

The scope of this chapter-formatted mini-series is to provide statistical tools for comparing two columns of data, X and Y. With respect to analytical applications such data may be represented for simple linear regression as the concentration of a sample (X) versus an instrument response when measuring the sample (Y). X and Y may also denote a comparison of the reference analytical results (X) versus predicted results (Y) from a calibrated instrument. At other times one may use X and Y to represent the instrument response (X) to a reference value (Y). Whatever data pairs one is comparing as X and Y, there are several statistical tools that are useful to assess the meaning of a change in... 

Subject variability High intraindividual variability in QTc values (circadian and seasonal variation law of regression to the mean) High interindividual variability in QTc values (males versus females) Unknown prevalence in the general population of subjects carrying silent mutations in the ion channels responsible for cardiac repolarization (these subjects have normal QTc value but reduced repolarization reserve) Variability in the individual metabolic capacity for a given drug... 

A point which may need emphasis, stated clearly in Hunter ( 2 ), is the precise interpretation of the confidence band about the predicted amount. This is important since without a clearly understood meaning, the interval will not be useful for assessing the precision of the predicted amounts or concentrations nor for comparing the results from various laboratories. Another reason the user of these methods must understand the interpretation is because increased precision can be achieved in at least two ways -by additional replication of the standards, which reduces the width of the confidence band about the regression line, and by performing multiple determinations on the unknowns, which reduces the width of the interval about the mean instrument response of the unknown. The interval for U is then given by... 

First, amount error estimations in Wegscheider s work were the result of only the response uncertainty with no regression (confidence band) uncertainty about the spline. His spline function knots were found from the means of the individual values at each level. Hence the spline exactly followed the points and there was no lack of fit in this method. Confidence intervals around spline functions have not been calculated in the past but are currently being explored ( 5 ). 

Statistical Treatment of Data. Each dose ratio point was the mean of three determinations. SrTly points between 20 and 80% on the dose response curves were utilized for the calculation of the dose ratio. After the dose was converted to a log value and the response to percent of the maximum, a best fit line was determined by linear regression. 

The elements of fhe vector y are the reference values of the response variable, used for building the model. The uncertainty on the coefficient estimation varies inversely with the determinant of the information matrix (X X) which, in the case of a unique predictor, corresponds to its variance. In multivariate cases, the determinant value depends on the variance of fhe predictors and on their intercorrelation a high correlation gives a small determinant of the information matrix, which means a big uncertainty on the coefficients, that is, unreliable regression results. 

---