Logistic regression models using Python Dr Juan Klopper 2022-04-13 | In this last video tutorial in my seminar series on the fundamentals of linear models, I take a look at logistic regression. Unlike linear regression, analysis of variance (ANOVA), and analysis of covariance (ANCOVA), the data type of the dependent variable is binary categorical.Please watch the first three tutorials in this seminar series. Topics discussed in these video tutorials are used in this tutorial.Linear regression youtu.be/6gtEfIbx6ko Analysis of variance youtu.be/yeRtgKtbUFU Analysis of covariance youtu.be/FhZB1oGVrYc In logistic regression we use numerical or categorical variables to calculate an estimate of the probability of one of the classes or levels of the binary categorical dependent variable.In this video lecture I talk about probability, odds, and odds ratios before we build three different logistic regression models. You will see just how easy it is to do in Python and how simple it is to understand logistic regression and how to interpret the results.00:00 Introduction19:25 Probability Odds and Odds Ratios24:54 Continuous numerical variable as independent variable41:06 Categorical variable as independent variable46:51 Continuous numerical and categorical variable as independent variables
Column space of a matrix using Python and sympy Dr Juan Klopper 2022-04-12 | In this video tutorial I discuss the column space of a matrix. I use the sympy package in python to explain the column space. Python and the symbolic python package sympy makes it easy to work with mathematical symbols. It is excellent at linear algebra.In the video I continue on from the previous video tutorial on the nullspace of a matrix. I explain what the column space is and go through examples that include all the types of matrices that you will come across. I also discuss what effect elementary row operations have on the column space. At the end of the tutorial I also show you how to prove that the column space is a vector subspace. 00:00 Introduction03:55 Review of the nullspace05:56 Square matrix with linear independence09:19 Square matrix with linear dependence17:01 Rectangular matrix with more columns than rows20:43 Rectangular matrix with more rows than columns and linear independence22:27 Rectangular matrix with more rows than columns and linear dependence24: 16 Proof that the column space is a vector subspace
Analysis of covariance using Python Dr Juan Klopper 2022-04-11 | This is the third video lecture in my seminar series on linear models. Here, I discuss analysis of covariance (ANCOVA). We combine what we have learned about linear regression and analysis of variance.In ANCOVA we have a categorical variable as independent variable and a continuous numerical variable as dependent variable. This is typical in a study design, where participants are assigned to the levels made up by the unique values of the independent variable. Other variable might also have an effect on the dependent variable. We might not be able to control for these with the study design. If such a variable is a continuous numerical variable, we term it a covariate. So, in ANCOVA we look at the effect of the categorical independent variable on the continuous numerical variable after correcting for the covariate.There are important assumptions for the use of ANCOVA models. The index below shows all the sections of the video lecture.Please watch the first two videos in this seminar series, as I use concepts explained in these two videos in the tutorial on ANCOVA.Linear regression: youtube.com/watch?v=6gtEfIbx6ko ANOVA: youtube.com/watch?v=yeRtgKtbUFU 00:00 Introduction05:55 Import data20:23 Linearity22:02 Homogeneity of regression 24:56 Normality of residuals26:43 Homogeneity of variance of the residuals27:53 Outliers29:16 ANCOVA
Nullspace of a matrix Homework with Python and sympy Dr Juan Klopper 2022-04-03 | In this video tutorial I discuss the nullspace of a matrix. I use the sympy package in python do explain the nullspace. Python and the symbolic python package sympy makes it easy to work with mathematical symbols. It is excellent at linear algebra.In the video I explain what the nullspace is and go through examples that include all the types of matrices that you will come across. I also discuss linear independence and make clear how the nullspace differs from the column space. At the end of the tutorial I also show you how to prove that the nullspace is a vector subspace. 00:00 Introduction05:59 Square matrix with linearly independent column vectors11:03 Square matrix with dependent column vector11:16 Rectangular matrix with more columns than rows with a single dependent column vector20:01 Rectangular matrix with more columns than rows and more than one dependent column vector21:48 Rectangular matrix with more rows than column and with independent column vectors24:39 Rectangular matrix with more rows than columns and a dependent column vector27:18 Proof that the nullspace is a subspace
Analysis of variance ANOVA using Python Dr Juan Klopper 2022-04-01 | In this second video tutorial in my seminar series on linear models, I discuss one-way analysis of variance (ANOVA). This video follows from the first tutorial on linear regression.In the video I use a Jupyter notebook in Visual Studio Code. I generate random data and walk through understandig the results of ANOVA by looking at the research equation, the null hypothesis and alternative hypothesis, the coefficients, standard errors, t statistics, and their p values, the confidence intervals for the coefficients, and then the analysis of variance table. The latter includes the degrees of freedom, the sum of squared errors (due to the regression and the error), the mean sum of squares, the F ratio, and it p value.Lastly, I take a look at post-hoc tests and use the Tukey HSD and Bonferroni methods.Here are time stamps for the various section in the tutorial.00:00 Introduction03:20 Generate data08:45 Exploratory analysis20:25 Generate ANOVA model22:52 Understanding the coefficients33:13 Understanding the errors41:04 Post-hoc tests
Linear Regression using Python Dr Juan Klopper 2022-03-25 | This seminar series looks at four important linear models (linear regression, analysis of variance, analysis of covariance, and logistic regression). A video that explains all four model types is at youtube.com/watch?v=SV9AxXFWZnM&t=12s This video is on linear regression using Python and the statsmodels package. In it, I demonstrate how to create data, and how to summarize and visualize the data for a simulated research project. I state the research question and the null hypothesis and alternative hypothesis.There is a thorough explanation of a linear model in one variable in order to build intuition about the results of linear regression. I explain what all the important results are and show how they are calculated. These include the parameters estimates (coefficients), the standard error, test statistics, p value, and the confidence intervals.I also explain the errors in the model such the sum of squared errors due to the regression and the residuals, the degrees of freedom in the model, the F statistic and its p value, as well as the coefficient of determination.
Linear regression ANOVA ANCOVA Logistic Regression Dr Juan Klopper 2022-03-15 | In this video tutorial you will learn about the fundamentals of linear modeling: linear regression, analysis of variance, analysis of covariance, and logistic regression. I work through the results of these tests on the white board, so no code and no complicated equations.Linear regression is the fundamental linear model. I dive deep into this simplest of linear models and look at a typical results table and what it all means. The other three models follow quite naturally from linear regression.00:00:00 Linear regression00:31:14 Analysis of variance ANOVA00:48:27 Analysis of covariance ANCOVA01:11:44 Logistic regression
Install Julia with Jupyter and Pluto notebooks and VS Code Dr Juan Klopper 2021-12-14 | #Julia #Jupyter #Pluto #VSCodeA quick video to take you through the installation of Julia, setting it up to run in the Terminal or Command Prompt, creating environments, setting up Jupyter notebooks and Pluto notebooks, and finally running Julia in Visual Studio code.00:00:00 Install Julia00:04:29 Running Julia from a command prompt or terminal00:08:03 Creating environmnets00:12:09 Jupyter notebooks00:14:38 Pluto notebooks and environments00:19:59 Visual Studio Code00:22:40 End
F ratio sampling distribution under the null hypothesis using Julia Dr Juan Klopper 2021-12-04 | In this video lecture I take a look at randomization under the null hypothesis to build a sampling distribution of F ratios (F statistics). I use this to compare the difference between three means.The chapter markers below give an indication of the content of the video. I use the Julia language in Visual Studio Code.00:00:00 Introduction00:00:43 Packages used in this notebook00:01:29 F distribution given two parameters00:02:47 Generating data for this notebook00:09:11 Summary statistics00:10:27 Visualizing the data00:11:00 Calculating the test statistic00:20:32 Randomization to create the sampling distribution under the null hypothesisThe file is available on GitHub at github.com/juanklopper/Julia-for-scientifc-computing/blob/master/FRatioSimulation.jl A PDF of the Pluto notebook is avaliable on GitHub at github.com/juanklopper/Julia-for-scientifc-computing/blob/master/FRatioUnderTheNullHypothesis.pdf #VSCode #Julia #Statistics #F #Distribution #statistic #Randomization #null #hypothesis
Estimating a p value for the difference in means between two groups Dr Juan Klopper 2021-12-02 | In this video tutorial I show you how to use Julia to estimate a p value for the difference in means for a numerical variable between two groups. This helps you to develop an inuitive understanding of expressing the likelihood of the difference that you find in your data.Under a simple null hypothesis there is no difference in the mean of a variable between two groups. This means that we can randomly reassign individuals / instances between the two groups and that should not change the means. If our test statistic of interest is the difference in means, we can do this random reassignment over and over again, each time recording our test statistic. This will build up a sampling distribution of the test statistic and we can compare our original test statistic with this sampling distribution.You can watch my video on installing environments for your Julia projects at youtube.com/watch?v=MC-wjQTXOOA The file is available at github.com/juanklopper/Julia-for-scientifc-computing/blob/master/ResamplingUnderTheNullHypothesis.jl There are chapter markers below if you want to skip ahead.00:00:00 Introduction00:00:28 Julia Language00:00:49 Visual Studio Code00:01:00 Terminal00:02:08 Lauch Visual Studio Code00:02:28 Julia Extension for Visual Studio Code00:03:22 Contents of the tutorial00:04:25 Julia packages used in this tutorial00:05:32 Generating random data00:08:23 Making sub-DataFrames00:09:48 Extracting vectors from a DataFrame00:11:06 Descriptive statistics00:12:05 Data visualization00:14:28 Test statistic00:15:14 Reassignment under the null hypothesis00:18:30 Visualize the sampling distribution00:20:06 Estimating a p value00:21:29 Verifying the p value00:21:54 End
Training a neural network in Keras with class imbalance Dr Juan Klopper 2021-09-19 | In this short video tutorial I explain the example page on the Keras website pertaining to training a binary classification network when there is an imbalance in the two classes of the target variable.
Deep Learning in Healthcare - New course announcement Dr Juan Klopper 2021-08-23 | @StellenboschUni This is an announcment video of my course at Stellenbsoch University on deep learning in healthcare.
Random Forests and Tensorflow Decision Forests - Video Lecture 14 Dr Juan Klopper 2021-08-21 | Ensemble techniques such as random forests have become very successful algorithms in machine learning. In this module we learn how the basic building block of a rando forest, called a decision tree, works. We also look at how to use Google’s brand-new Decision Forest module for TensorFlow. You can watch the video or read the PDF document. There is no exercise file to complete.
k Nearest Neighbors - Video Lecture 13 Dr Juan Klopper 2021-08-21 | In this module we work through the typical process of a machine learning project using the k nearest neighbours algorithm. You can watch the video lecture or read the PDF document. There is no exercise file to complete.
Linear Modeling using Statsmodels - Video Lecture 11 Dr Juan Klopper 2021-08-21 | This module aims to explain as much of the basics of linear modelling as possible. We start with correlation and then learn about linear regression. You can watch the video lecture or read the PDF document. There is an exercise file that you should attempt to complete.
Uncertainty and Bootstrap Confidence Intervals - Video Lecture 10 Dr Juan Klopper 2021-08-21 | Our results are invariably from a sample taken from a larger population. As such, there is uncertainty in our findings. In this module we learn to express this uncertainty as confidence intervals using bootstrap resampling. You can watch the video lecture or read the PDF document.
Comparing Means - Video Lecture 09 Dr Juan Klopper 2021-08-21 | In this module we put our knowledge from the previous module to work, comparing two means for a numerical variable. We learn how to use the null hypothesis to construct many possible outcomes and to see the probability of the findings from our won research. There is an accompanying video lecture that you can watch, or you can read the PDF document. Also attempt to complete the exercise file.
Randomness and Sampling - Video Lecture 07 Dr Juan Klopper 2021-08-16 | Data captures values for random events. It is important to have a basic understanding of the concept of randomness and the probability of findings from data analysis. In this module we also explore the patterns of data values called distribution. Watch the video lecture or read the PDF document. There is an exercise file that you should attempt to complete.
Data Visualisation using Plotly - Video Lecture 06 Dr Juan Klopper 2021-08-16 | The cornerstone of Data Science is data visualisation. The ability to plot our data gives us a rich understanding of the data and allows us to communicate our findings. You can watch the video lecture or read the PDF document. There is an exercise file that you should attempt to complete.
Hypothesis Testing - Video Lecture 08 Dr Juan Klopper 2021-08-16 | Our scientific method, the steps that we follow to understand our world through research, is based on the process of hypothesis testing. You can watch the video lecture or read the PDF to learn more. There is no exercise file for this module.
Logistics For the Stellenbosch Uiversity Course on R for healthcare statistics Dr Juan Klopper 2021-08-11 | This is a short follow-up to the announcement video on the new Stellenbosch University short course on R for healthcare and life science statistics starting next week. It contains some logistics for those that are, or have, signed up.The link to sign up is at at https://shortcourses.sun.ac.za/application-form.html?offeringid=6cc261b2-45da-eb11-abe4-005056801c40To dowload R go to cran.r-project.org and follow the links for your operating system.Once you have installed R, go to rstudio.com/products/rstudio/download to dowload and install RStudio, the coding environment that we will use during this course. All the rquired study files will be send to those that sign up.You can also run RStudio in the cloud. Sign up for a free account at https://rstudio.cloud If you want to go this route, a link will be sent to you after signing up to this course. The email link will take you to your RStudio account in the cloud (make sure that you are signed in) and provide a copy of all the study files.The following R packages are used in the course:data.tableDTtidyverseknitrcargmodelsLooking forward to seeing you.
Summary Statistics - Video Lecture 05 Dr Juan Klopper 2021-08-11 | Module five introduces the descriptive statistics. Describing our data is the first step in understanding the story that our data is trying to tell. You may either watch the video lecture or read the PDF document. There is an exercise notebook that you should complete.
Pandas for data wrangling - Video Lecture 04 Dr Juan Klopper 2021-08-03 | In this module we import data from spreadsheet and learn how to manipulate it using one of the most successful packages called pandas. You can watch the video lecture or read the PDF document. There is an exercise file that you should attempt to complete.
Introduction to Python - Video Lecture 03 Dr Juan Klopper 2021-08-03 | This module introduces the Python programming language. Watch the accompanying video lecture or read the PDF document. There is an exercise file that you should attempt to complete. A new computer language can only be conquered by using it yourself.
Introduction to Data Science - Video Lecture 01 Dr Juan Klopper 2021-08-03 | This module introduces the topic of Data Science and the tools that we use to do our analysis. At the end of the module there is a short example project.
Announcing R for Statistics Course Dr Juan Klopper 2021-08-03 | @R @statistics @StellenboschUni This is an announcement of my course for the School of Data Science and Computational Thinking at Stellenbosch University.The course starts on the 16th of August 2021. You will find a link to the course and information on signup at https://shortcourses.sun.ac.za/courses/c-19/2021-5760.html This course aims to teach the fundamentals of statistics for helathcare and the life sciences using the R language.There are 15 modules in the course. All are summarised in the video.
Matplotlib tutorial for plotting in Python Dr Juan Klopper 2021-07-23 | #Python #Matplotlib #Jupyter #PlottingMatplotlib is the grandparent of plotting libraries in Python. There is not much that you cannot plot with this package. That makes it daunting to use.In 80 minutes I try to give you the best introduction to using matplotlib for scientific plots.0:00:00 I ramble about the tutorial :)0:01:15 Tutorial starts0:02:40 Table of contents0:4:17 Python packages used in this tutorial0:06:24 Pyplot interface0:45:41 Object oriented interface1:02:27 Plots for functions and vectors1:18:58 Saving plotsWATCH NEXT:▶️ Linear algebra using sympy: youtu.be/MBhPaqKRehg▶️ Gram Schmidt process for QR decomposition using Python: youtu.be/fdshjUzUWTs//You Might Also Like:~~~~~~~~~~~~~~~Udemy: udemy.com/mathematica-for-statistics Udemy: udemy.com/biostatistics-fundamentals-using-python Udemy: udemy.com/mathematica Udemy: udemy.com/spss-for-healthcare-and-life-science-statistics RPubs: http://rpubs.com/juanhklopper Github: github.com/juanklopper?tab=repositories//CONNECT WITH ME~~~~~~~~~~~~~~~👉 Web: http://www.juanklopper.com
Want to learn Data Science? Announcing a New Course from Stellebosch University Dr Juan Klopper 2021-07-18 | #DataScience #Python #Course #Stellenbosch #University #Google #ColabThe signup link is here: http://www.sun.ac.za/english/data-science-and-computational-thinking/Pages/Introduction-to-Data-Science-and-Computational-Thinking.aspx This is an announcement about a new course in Data Science from Stellenbosch University.I am Dr Juan Klopper from the School for Data Science and Computational Thinking. Hundreds of participants have already signed up. Check back right here in the description for full details on worldwide sign up and further notices.In this course I teach Data Science using Python. The course will be easy to follow with video lectures, PDF documentation, exercise files, and live online sessions.During the course we will use Google Colab notebooks so that no one has to worry about any local installations.You will learn about Python, working with data, summary statistics, data visualisation, probabilities, distributions and sampling distributions, bootstrap resampling, uncertainty, confidence intervals, comparison of means, linear models, nonparametric tests, and comparing categorical variables. You will also learn about machine learning techniques such as k nearest neighbours and random forests and I will show you how to construct a machine learning project.Watch the video for more information and tell everyone you know. Leave comments down below if you have questions. I hope to see you on the course.SCHEDULEWeek 101 Data Science02 Data and Definitions03 The Python Language Week 204 Importing and Manipulating Data05 Descriptive Statistics06 Data Visualisation Week 307 Randomness and Sampling08 Hypothesis Testing09 Comparing Means Week 410 Uncertainty11 Linear Models Week 512 Machine Learning13 k Nearest Neighbours14 Random Forests
Bootstrap resampling for confidence intervals and sampling distrubutions Dr Juan Klopper 2021-07-09 | #DataScience #Bootstrap #Resampling #Julia #VSCode #Confidence #Intervals #ProbabilityIn this video I show you how to think about using bootstrap resampling to express uncertainty in a statistic and also how to think about the sampling distribution of a test statistic and the probability of that test statistic. Using the Julia Language in Visual Studio Code we generate values for a continuous numerical variable for two groups (generated from the two sample space elements of a categorical variable).Our test statistic is the difference in means for the variable between the two groups. We generate a sampling distribution of the test statistic under the assumption of the null hypothesis. The latter states that there is no difference between the two means, that is, the test statistic is zero. Given this assumption we can randomly reassign subjects to the two groups and recalculate the test statistic. Doing this many times over produces the sampling distribution. We can then see how many of the resampled tests statistics are below the negative of our test statistic and how many are more than positive our test statistic. Adding these and expressing it as a fraction of the number of resamples gives us the p value for our statistic.We also express the mean for the variable in both groups as a whole. We must consider the uncertainty in this statistic. We estimate the lower and upper bounds for the parameter in the population. This is the confidence interval.
Bootstrap resampling for probability and confidence intervals Dr Juan Klopper 2021-07-09 | #DataScience #Bootstrap #Resampling #Wolfram #Confidence #Intervals #ProbabilityIn this video I show you how to think about using bootstrap resampling to express uncertainty in a statistic and also how to think about the sampling distribution of a test statistic and the probability of that test statistic.Using the Wolfram Language we generate values for a continuous numerical variable for two groups (generated from the two sample space elements of a categorical variable).Our test statistic is the difference in means for the variable between the two groups. We generate a sampling distribution of the test statistic under the assumption of the null hypothesis. The latter states that there is no difference between the two means, that is, the test statistic is zero. Given this assumption we can randomly reassign subjects to the two groups and recalculate the test statistic. Dong this many times over produces the sampling distribution. We can then see how many of the resampled tests statistics are below the negative of our test statistic and how many are more than positive our test statistic. Adding these and expressing it as a fraction of the number of resamples gives us the p value for our statistic.We also express the mean for the variable in both groups as a whole. We must consider the uncertainty in this statistic. We estimate the lower and upper bounds for the parameter in the population. This is the confidence interval.
Basics of the Jacobian and its use in a neural network using Python Dr Juan Klopper 2021-06-22 | #Python #DataScienceIn this 20 minute video I introduce the topic of the the Jacobian. It is simply a matrix of partial derivatives of single and multivariable functions or vector valued functions.While the Jacobian is easy to calculate by hand, we can also use the symbolic Python package sympy to verify our work.The Jacbian has many uses. In this video I take a look at its use in the forward propagation step of a simple neural network.
Gram-Schmidt orthogonalization and Legendre Polynomials using the Wolfram Language Dr Juan Klopper 2021-05-12 | #LinearAlgebra #DataScienceIn this video tutorial I show you how to orthogonalize a basis for a vector space of polynomials. It might seems a strange thought, but yes, polynomials that are orthogonal to each other can be created.In the tutorial I explain the concept of the inner product between two polynomials and use the Gram-Schmidt process to create the orthogonal set of polynomials.While it is an easy task, I use the Wolfram Language in this video to make life (and understanding) even easier. Code along if you have access to the Wolfram Language or search Google for "Wolfram Programming Lab" and sign up for a free account. You can then code along right in your favorite browser.At the end of the tutorial, when we have our set of orthogonal polynomials, I also discuss how they are slightly modified to give us the Legendre polynomials.
Gram Schmidt process for QR decomposition using Python Dr Juan Klopper 2021-05-12 | #LinearAlgebta #DataScienceIn this video tutorial I use Python to explain the easy steps of the Gram Schmidt process. Following the steps of this process yields a set of orthonormal basis vectors for the (sub) space spanned by the column vectors of a matrix. Each new column vector is orthogonal to all the other column vectors and each is of unit length.This change of basis is used in QR factorization or QR decomposition of a matrix.WATCH NEXT:▶️ Modern Linear Algebra using Python: youtu.be/MBhPaqKRehg▶️ Second order ODEs solved with Python using the Euler formula: youtu.be/pV0SY0EtKaI▶️ Gram Schmidt process for QR decomposition using Python: youtu.be/fdshjUzUWTs▶️ Second order ODE solved with RK4 in Python: youtu.be/-k64Xa0onfQ▶️ Solving a system of first order ODEs with RK4 using Python: youtu.be/cQSQCv0XBx8//You Might Also Like:~~~~~~~~~~~~~~~Udemy: udemy.com/mathematica-for-statistics Udemy: udemy.com/biostatistics-fundamentals-using-python Udemy: udemy.com/mathematica Udemy: udemy.com/spss-for-healthcare-and-life-science-statistics RPubs: http://rpubs.com/juanhklopper Github: github.com/juanklopper?tab=repositories//CONNECT WITH ME~~~~~~~~~~~~~~~👉 Web: http://www.juanklopper.com
Gram Schmidt process for QR decomposition using Julia Dr Juan Klopper 2021-05-12 | In this video tutorial I use the Julia Language to explain the easy steps of the Gram Schmidt process. Following the steps of this process yields a set of orthonormal basis vectors for the (sub) space spanned by the column vectors of a matrix. Each new column vector is orthogonal to all the other column vectors and each is of unit length.This change of basis is used in QR factorization or QR decomposition of a matrix.If you want to learn more about the Julia Language take a look at the Playlists here on my channel or earn a certificate from my massive open online course on Coursera coursera.org/learn/julia-programming.
Ordinary least squares tutorial using Julia Dr Juan Klopper 2021-05-08 | In this video tutorial I develop an inituitive understanding of ordinary least squares (OLS) and how it pertains to finding a quadratic equation, a cubic equation, and a linear equation given three point in the plane.Developing an intuition for OLS requires understanding of the column space of a matrix and how to project a vector onto that space. This is done through a simple schematic and I use Julia code to show just how easy it is to calculate using ordinary least squares.
Ordinary Least Squares Tutorial using Python Dr Juan Klopper 2021-05-08 | In this video tutorial I discuss the creation of a quadratic, a cubic, and a linear equation given three points in the plane. These approximations (solution in the case of the cubic equation) can be calculated using ordinary least squares (OLS).In the video I develop and intuitive and visual explanation of OLS as a projection onto the column space of a matrix. The calculations are done using Python.
Linear models using the F distribution in python Dr Juan Klopper 2021-05-01 | In this video tutorial I review the use of the F distribution in linear regression, the t test, and in analysis of variance (ANOVA).The turotial includes:An overview of the F distribution given two parametersAssessment of a linear regression mode using the coefficient of determinationCalculating an F statistic in linear regressionCalculating a p value from the F statisticThe F statistic in the t test and calculating a p value for the statisticThe F statistic in ANOVA and calculating a p value for the statisticThe fundamentals are introduced on the white board, before we open a Jupyter notebook and use python to create an intuitive understanding of the use of the F distribution.
All the planets fit between the earth and the moon (sometimes) Dr Juan Klopper 2021-04-26 | #Wolfram #MathematicaIn this fun tutorial I use the Wolfram Langue to look at which planets fit in the distance between the earth and the moon. The Wolfram Language has all of the data built-in.
Design Matrices using Patsy in Python Dr Juan Klopper 2021-04-24 | I am Dr Juan Klopper, Surgeon and Research Fellow as the School for Data Science and Computational Thinking at Stellenbosch University. If you want to learn how to use python for healthcare statistics consider, my course and get a certificate from my research group. The course is available at udemy.com/course/biostatistics-fundamentals-using-python For more course visit juanklopper.comThe files for this tutorial will be made available on GitHub soon.In this video tutorial I discuss the preparation of tabular data for use in statistical python packages such as statsmodels. The patsy package is ideal for creating design matrices that can be used in statsmodels.Design matrices in models include the dependent vector and the matrix of independent variables. Functions in patsy can generate these from tabular data. The patsy package can also preprocess the data. It can, for instance, generate design matrices with interaction terms, add new variables to a feature matrix, transform the values of a variable, and many other such transformation.
Relative risk and efficacy using AstraZeneca vaccine trial Dr Juan Klopper 2021-03-07 | The aim of new treatment interventions such as vaccines, is to lower the risk of disease. This is a very topical discussion as many vaccines have been, or are currently, subjected to clinical trials.These trial express their results as efficacy. Efficacy is calculated from relative risk. It is hoped that a vaccine will lower the risk of infection when compared to placebo.In this video tutorial, I use the R language for statistical computing to explain risk, relative risk, efficacy, and most importantly, confidence intervals for efficacy.In February 2021 South Africa received their first batch of vaccines (Astra Zeneca). Prior to a roll out to healthcare workers a small trial appeared which showed efficacy of only 21%. For this, and other findings in the trial, the vaccine was withdrawn.Because the trial was so small and the risk of developing mild and moderate COVID-19 was low in each group, the confidence intervals around the efficacy was very wide.Follow along in this video to learn more about risk, realative risk, efficacy and confidence intervals. The RPubs documents are avalaible at rpubs.com/juanhklopper and the R markdown files are available at github.com/juanklopper/R_statistics Please remember to like this video. Subscribe and leave a comment.
Calculus of polar coordinate functions Dr Juan Klopper 2021-03-04 | This video contains 16 problems pertaining the derivative and the integral of polar coordinate functions. Problem 1 in the first section derives an equation for the derivative and problem 1 in the second section derives an equation for the integral of a function expressed in polar coordinates.Problem 1.1 - 00:00:37Problem 1.2 - 00:04:27Problem 1.3 - 00:08:57Problem 1.4 - 00:15:40Problem 1.5 - 00:20:22Problem 1.6 - 00:22:43Problem 2.1 - 00:34:18Problem 2.2 - 00:41:06Problem 2.3 - 00:49:17Problem 2.4 - 00:54:28Problem 2.5 - 01:01:25Problem 2.6 - 01:05:31Problem 2.7 - 01:12:19Problem 2.8 - 01:17:21Problem 2.9 - 01:21:04Problem 2.10 - 01:26:23If you want to get the questions in PDF format to try them yourself, visit github.com/juanklopper/Vector-calculus . There is also a PDF files containing notes on the calculus of polar coordinates, as well as all the other files in the series. Remember to check out the Wolfram Language code to try it yourself.
Polar and rectangular coordinate equations Dr Juan Klopper 2021-02-06 | This is the fourth problem set in this new series on vector calculus. In this video lecture, we look at polar coordinates and the curves that result from polar equations. We start by converting between rectangular and polar coordinates and also look at converting between rectangular coordinate and polar coordinate equations.The problem sets for this series are at github.com/juanklopper/Vector-calculus and the one specific to this video is at github.com/juanklopper/Vector-calculus/blob/main/04%20Polar%20coordinates.pdf The time codes for all the problems are listed below.Problem 1.1 – 00:00:14Problem 1.2 – 00:02:21Problem 1.3 – 00:03:43Problem 1.4 – 00:13:16Problem 1.5 – 00:16:05Problem 1.6 – 00:18:02Problem 1.7 – 00:19:57Problem 2.1 – 00:21:34Problem 2.2 – 00:23:36Problem 2.3 – 00:24:32Problem 2.4 – 00:26:57Problem 2.5 – 00:27:43Problem 2.6 – 00:29:41Problem 2.7 – 00:31:38Problem 2.8 – 00:35:31Problem 2.9 – 00:38:05Problem 2.10 – 00:41:32Problem 2.11 – 00:42:03Problem 3.1 – 00:42:42Problem 3.2 – 00:44:46Problem 3.3 – 00:46:33Problem 3.4 – 00:49:12Problem 3.5 – 00:50:28Problem 3.6 – 00:52:01Problem 3.7 – 00:53:35Problem 3.8 – 00:56:10Problem 3.9 – 00:59:00Problem 4.1- 01:00:55Problem 4.2 – 01:06:30
Arc length of a parametric curve Dr Juan Klopper 2021-02-02 | In this third video in the series on vector calculus, I take you through example probelms involving arc length of a parametric curve. The problem set is avalaible as a PDF file at github.com/juanklopper/Vector-calculus if you want to try them before watching this tutorial.The time codes for the problems are listed below.Problem 1 – 0:55Problem 2 – 13:17Problem 3 – 16:23Problem 4 – 20:15Problem 5 – 25:12
Calculus of parametric equations Dr Juan Klopper 2021-02-02 | In this video, we take a look at the first and second derivatives of parametric equations. The first two questions relate to deriving equations for the derivatives. We also take a look at calculating the area enclosed by a parametric curve using integration. The problems are available as a PDF file at github.com/juanklopper/Vector-calculus if you want to try them before watching the tutorial.The time code for the problems are listed below.Problem 1.1 - 0:00:29Problem 1.2 – 0:04:07Problem 1.3 – 0:07:48Problem 1.4 – 0:10:14Problem 1.5 – 0:16:16Problem 1.6 – 0:17:30Problem 1.7 – 0:23:28Problem 1.8 – 0:24:39Problem 1.9 – 0:27:27Problem 1.10 – 0:32:20Problem 1.11 – 0:34:22Problem 1.12 – 0:36:39Problem 1.13 – 0:39:24Problem 1.14 – 0:47:59Problem 2.1 – 0:54:48Problem 2.2 – 0:58:05Problem 2.3 – 1:04:17Problem 2.4 – 1:09:25Problem 2.5 – 1:14:10Problem 2.6 – 1:16:24
Introduction to parametric equations Dr Juan Klopper 2021-01-20 | This new series is all about vector calculus through solving example problems. Find the problem set for this video lecture at github.com/juanklopper/Vector-calculus I do some of the graphing using the Wolfram Language. Code along for free on the eb by creating an account at wolframcloud.com
Tutorial on matrix algebra with help from the Julia language Dr Juan Klopper 2021-01-08 | This is part 3 of the section on matrices with a little help from Julia. Matrices are mathematical objects and in this video we add and subtract matrices and multiply a matrix with a scalar.To learn how to set up a Julia environment and use it in a Pluto notebook watch this: youtube.com/watch?v=6ewFgLR1LLY&t=35s
Solving linear systems using elementary row operations with Julia Dr Juan Klopper 2021-01-08 | This is part 2 on the second section on matrices. In this video I look at linear systems and how to solve them by hand using elementary row operations.It is much easier to do using Julia, though!To set up a Julia environment and use it in Pluto notebooks, watch this: youtube.com/watch?v=6ewFgLR1LLY&t=35s
Tutorial on matrices with help from Julia Dr Juan Klopper 2021-01-08 | This is the second part of my video series on linear algebra with a little help from the Julia Programming Language.The first video was on vectors. Since there is so much to learn about matrices, this second section will be split into nine videos.This first video simply shows you what a matrix is and how we denote it.To learn how to set up a Julia environment and to use it in a Pluto notebook, watch this video: youtube.com/watch?v=6ewFgLR1LLY&t=35s
How to extract COVID-19 info from JHU data using R Dr Juan Klopper 2020-12-30 | In this short video I show you one way to extract data from the John Hopkins University data repository on coronavirus.I look at data specific to South Africa and contrast it to data from Germany and the United Kingdom. Since South Africa is behind these countries in COVID-19, we can use data from these nation in helping to predict what will happen in South Africa at the start of 2021.
Data transformation using the Wolfram Language Dr Juan Klopper 2020-12-30 | In this short video I show you how to transform data for use in parametric tests. When data does not meet the assumption of normality, we can transform it using non-linear functions.I state, though, that it is probably better to use distribution-independent (non-parametric) tests.In the video, we generate data that not normally distributed and use the log and square root transformations to transform it.The file is available at github.com/juanklopper/Coursera-Doing-clinical-research and at wolframcloud.com/obj/juan.klopper/Published/Data%20transformation%20for%20statistical%20analysis.nb
