The Experts below are selected from a list of 234 Experts worldwide ranked by ideXlab platform
Roland Fried - One of the best experts on this subject based on the ideXlab platform.
-
Computing the update of the repeated median Regression Line in Linear time
Information Processing Letters, 2003Co-Authors: Thorsten Bernholt, Roland FriedAbstract:The repeated median Line estimator is a highly robust method for fitting a Regression Line to a set of n data points in the plane. In this paper, we consider the problem of updating the estimate after a point is removed from or added to the data set. This problem occurs, e.g., in statistical onLine monitoring, where the computational effort is often critical. We present a deterministic algorithm for the update working in O(n) time and O(n2) space.
F. Xavier Rius - One of the best experts on this subject based on the ideXlab platform.
-
Robust Linear Regression taking into account errors in the predictor and response variables.
The Analyst, 2001Co-Authors: F. Javier Del Río, Jordi Riu, F. Xavier RiusAbstract:We developed a robust Regression technique that is a generalization of the least median of squares (LMS) technique to the field in which the errors in both the predictor and the response variables are taken into account. This simple generalization is limited in the sense that the resulting straight Line is found by using only two points from the initial data set. In this way a simulation step is added by using the Monte Carlo method to generate the best robust Regression Line. We call this new technique ‘bivariate least median of squares’ (BLMS), following the notation of the LMS method. We checked the robustness of the new Regression technique by calculating its breakdown point, which was 50%. This confirms the robustness of the BLMS Regression Line. In order to show its applicability to the chemical field we tested it on simulated data sets and real data sets with outliers. The BLMS robust Regression Line was not affected by many types of outlying points in the data sets.
-
Lack of fit in Linear Regression considering errors in both axes
Chemometrics and Intelligent Laboratory Systems, 2000Co-Authors: Àngel Martı́nez, Jordi Riu, F. Xavier RiusAbstract:Abstract Testing for lack of fit of the experimental points to the Regression Line is an important step in Linear Regression. When lack of fit exists, standard deviations for both Regression Line coefficients are overestimated, and this gives rise, for instance, to confidence intervals that are too large. If these confidence intervals are then used in hypothesis tests, bias may not be detected so there is a greater probability of committing a β error. In this paper, we present a statistical test, which analyses the variance of the residuals from the Regression Line whenever the data to be handled have errors in both axes. The theoretical expressions developed were validated by applying the Monte Carlo simulation method to two real and nine simulated data sets. Two other real data sets were used to provide examples of application.
Lutgarde Thijs - One of the best experts on this subject based on the ideXlab platform.
-
Response to Determinants of the Ambulatory Arterial Stiffness Index Regression Line
Hypertension, 2009Co-Authors: Ahmet Adiyaman, Dirk G. Dechering, Theo Thien, José Boggia, Ji-guang Wang, Tine W. Hansen, Eoin O'brien, Tom Richart, Lutgarde ThijsAbstract:We thank Schillaci and colleagues for their continued interest1–3 in the ambulatory arterial stiffness index (AASI). Schillaci et al1 reported that the inverse association between AASI and nocturnal dipping is stronger for diastolic than for systolic blood pressure. This observation has no repercussion on r 2, which is a measure of fit of the Regression Line. When A (diastolic blood pressure) is regressed on B (systolic blood pressure) or vice versa, estimates of r 2 are exactly the same. With regard to the proposed threshold value of r 2 (0.36), Schillaci and colleagues …
Thorsten Bernholt - One of the best experts on this subject based on the ideXlab platform.
-
Computing the update of the repeated median Regression Line in Linear time
Information Processing Letters, 2003Co-Authors: Thorsten Bernholt, Roland FriedAbstract:The repeated median Line estimator is a highly robust method for fitting a Regression Line to a set of n data points in the plane. In this paper, we consider the problem of updating the estimate after a point is removed from or added to the data set. This problem occurs, e.g., in statistical onLine monitoring, where the computational effort is often critical. We present a deterministic algorithm for the update working in O(n) time and O(n2) space.
Ahmet Adiyaman - One of the best experts on this subject based on the ideXlab platform.
-
Response to Determinants of the Ambulatory Arterial Stiffness Index Regression Line
Hypertension, 2009Co-Authors: Ahmet Adiyaman, Dirk G. Dechering, Theo Thien, José Boggia, Ji-guang Wang, Tine W. Hansen, Eoin O'brien, Tom Richart, Lutgarde ThijsAbstract:We thank Schillaci and colleagues for their continued interest1–3 in the ambulatory arterial stiffness index (AASI). Schillaci et al1 reported that the inverse association between AASI and nocturnal dipping is stronger for diastolic than for systolic blood pressure. This observation has no repercussion on r 2, which is a measure of fit of the Regression Line. When A (diastolic blood pressure) is regressed on B (systolic blood pressure) or vice versa, estimates of r 2 are exactly the same. With regard to the proposed threshold value of r 2 (0.36), Schillaci and colleagues …
