What is a good sample size? How many replicate measurements do we need to make inferences about a population from the sample? There are scientific articles on this subject, such as the one by H.W. Austin (1983), of which the title of the blog post is borrowed. There is no universal answer to this question. It depends very much on the studied phenomenon and the requirements on the results. Here is a nice example of how MATLAB helps to get a sense of the relationship between sample size and quality of the result.
MathWorks supports the curriculum development of the MATLAB/LEGO MINDSTORMS Environmental Remote Sensing Lab by Martin H. Trauth, in collaboration with Jens Tronicke and Bodo Bookhagen. Sebastian Gross, Senior Education Technical Specialist of MathWorks, advises us on all matters involving the development of the new curriculum. The course for undergraduate and graduate students aims to improve their skills to build efficient teams to solve typical problems in earth sciences in acquiring, processing and analyzing typical multispectral, geophysical and geometric data. The progress of course development as well as course materials will be provided on this blog.
Lisa Kempler, MATLAB Community Strategist at MathWorks, runs a great website at the Science Education Resource Center (SERC) at Carleton College, Northfield, Minnesota USA. The website provides a collection of geoscience-related MATLAB applications. Continue reading “SERC Teaching Geoscience with MATLAB”
Pearson’s linear product-moment correlation coefficient is highly sensitive to outliers, as can be illustrated by the following example. Several alternatives exist, such as Spearman’s rank correlation coefficient and the Kendall’s tau rank correlation coefficient, both contained in the Statistics and Machine Learning Toolbox. Continue reading “Outliers and Correlation Coefficients”
Since the introduction of electronic devices with touch controls, interactive 3D graphics objects have become increasingly popular in multimedia electronic books (ebooks). The Simulink 3D Animation available from The MathWorks, Inc. provides the necessary tools to create and export 3D graphics objects for inclusion in documents such as multimedia ebooks, interactive webpages, and presentations. Continue reading “Creating Animated 3D Files with MATLAB”
We can create video files, which we then integrate into multimedia publications. In principle, any video material can be used, such as recorded laboratory and field experiments, screen video captures or recorded computer animations. Here I demonstrate how to create a video of an animation with MATLAB.
Let us create some audio files with MATLAB, which we then integrate into multimedia publications. In principle, any audio material can be used, such as recorded nature sounds. Here is a preview of a section of a new chapter about multimedia publications of the upcoming 2nd Edition of MDRES.
A common error in the regression analysis is that bivariate data with an exponential relationship are log-transformed, then a best-fit line is calculated using a classical linear regression using the least squares, before the result is backtransformed. Here is a MATLAB example to show how to do it better. Continue reading “Classical Linear Regression of Log-Transformed Data”