MATLAB: Statistical Analysis in MATLAB

Name: MATLAB: Statistical Analysis in MATLAB
Author: Gabriel Popoola

Understand your data at a statistical level

Preview the first lesson free — get full access to all 3 lessons.

Course: On-Demand

Hello, my name is Gabriel Popoola and this is the statistical analysis in MATLAB module. During this module, we're going to cover summary statistics, descriptive statistics, and probability distributions. During this lesson I'm gonna show you two ways to view summary statistics for your data, first on the individual level, and then second using the grouping method. So the first thing we're gonna do is load our data set. And in this case it's gonna be the patients data set. And then the next thing we're gonna do is store our systolic and diastolic blood pressures into a single variable called blood pressure. So that's just gonna be systolic diastolic. And then we're going to create a data table, and our columns are going to be gender, age, blood pressure. And we're just gonna print our table here and we have gender, age, smoker, and blood pressure. So in order to get the summary, we simply just need to use the summary function T, and what you get is the minimum, the median, the maximum, and the breakdown for each feature if it's a Boolean. So for age, we have a minimum value of 25, max of 50, and a median of 39. For the smoker, which is logical or a Boolean, we have 34 trues and 66 falses. And then for blood pressure, we have our continuous values. And once again, we have a minimum, median, and a maximum. And then for gender, of course, because it's either gonna be male or female, and there aren't dummy variables, it's just gonna be different strings that you're working with here. And so this is basically how you get your summary statistics on the individual level in terms of features or columns. Next up I'm going to show you how to get your statistics based on groupings. And so to do that we're going to use the groupstats function. And so first I'm just going show you what groupstats does at a high level. So you pass in your table T and what you get is a pretty basic summary that gives you the number of elements. And so here we have the number of elements in each column being 100. And as you can probably tell this, isn't very helpful. However, what we can do here is group our data by gender, for instance. And so we run this groupstats T, but then we pass in this second parameter, gender. And now what you get is a breakdown of the remaining columns based on the gender. And so for gender, we have two, male and female. And it gives us the mean age, the mean smoker, and mean blood pressure for each gender. And so, as you know, the mean smoker or rather the smoker column is a logical column, meaning that the values are either true or false. And in this case, true is equal to one and false is equal to zero. And so that's how these means are calculated. And so, as another example we're gonna run groupstats again for T but this time we're going to group based on smoker. And once again, we have our categories, false and true. But this gives us the number of elements in each category. And so for gender, we have 66 that don't smoke and 34 that do. And because these are counts, what you see is that the number of true and false is the same for each one because this gives us groupings. And the reason why this gives us the number of elements instead of the mean is because we have this gender column here, and that limits what sort of summary statistics can be given since that's a string column. But if we drop that column, so if we create a table T2 and set that equal to age, blood pressure, and smoker, and then we get our groupstats again, T2 and smoker, and now we get our mean age and mean blood pressures. And so as a side by side comparison, we have our table T and our table T2, both grouped by smoker. And once we drop this categorical variable, we see that we get our means for our columns instead of just the element count, which in most cases isn't going to be very useful for datasets such as this. This concludes this lesson, thank you.

Advanced Provider : Gabriel Popoola 3 Lessons · 20m in Arabic, German, English, Spanish, French, Portuguese, Chinese

Course Description

Statistical analysis is one of the keys and foundational forms of analysis when it comes to gaining insights into data. Learning about new ways to calculate statistical values and represent your data using these values can be vital in genuinely understanding the nature of your data. I'm going to show you two ways to view summary statistics for your data. First, on the individual level, and second, using the grouping method. You’ll get an overview of descriptive statistics and some key functions in MATLAB. Descriptive statistics are brief descriptive coefficients that summarize a given dataset, which can represent the entire population or a sample of a population. They're broken down into two measures of central tendency and measures of variability or spread. In this MATLAB module, you will discover the usefulness and ability to gain statistical insights into your data using MATLAB.

After completing this module, you can apply what you learned to any project or task involving data/statistical analysis. This includes anything involving a dataset that contains numerical and/or categorical data. This knowledge will allow you to gain more insights into how you can use your data.

What You'll Learn

Recognize the difference between summary statistics and descriptive statistics
Generate group and individual statistics for your data in MATLAB
Calculate descriptive statistics, including measures of central tendency and measures of variability or spread
Define what probability distributions are and how they are used
Apply statistical analysis in MATLAB to projects involving numerical and/or categorical data

Key Takeaways

Statistical analysis is a foundational form of analysis for gaining insights into data.
Descriptive statistics are brief descriptive coefficients that summarize a dataset, which can represent an entire population or a sample of a population.
Descriptive statistics break down into two groups: measures of central tendency and measures of variability or spread.
Summary statistics for data can be viewed on an individual level and by using the grouping method.
The skills learned apply to any project or task involving a dataset that contains numerical and/or categorical data.

Frequently Asked Questions

What will I learn in this MATLAB module?

You will get an overview of descriptive statistics and some key functions in MATLAB, learn two ways to view summary statistics (on the individual level and using the grouping method), and learn what probability distributions are and how they are used.

Who is this course for?

It is for anyone who wants to gain statistical insights into their data using MATLAB and apply this to projects or tasks involving data and statistical analysis, including any dataset containing numerical and/or categorical data.

What is the difference between summary statistics and descriptive statistics?

The course teaches you to recognize the difference between the two; descriptive statistics are brief descriptive coefficients that summarize a given dataset, broken down into measures of central tendency and measures of variability or spread.

What topics does the course cover?

The course covers three lessons: Summary Statistics, Descriptive Statistics, and Probability Distributions.

What can I do after completing this module?

After completing this module, you can apply what you learned to any project or task involving data and statistical analysis, including anything involving a dataset that contains numerical and/or categorical data.