#### Overview of this book

Want to handle everything that Julia can throw at you and get the most of it every day? This practical guide to programming with Julia for performing numerical computation will make you more productive and able work with data more efficiently. The book starts with the main features of Julia to help you quickly refresh your knowledge of functions, modules, and arrays. We’ll also show you how to utilize the Julia language to identify, retrieve, and transform data sets so you can perform data analysis and data manipulation. Later on, you’ll see how to optimize data science programs with parallel computing and memory allocation. You’ll get familiar with the concepts of package development and networking to solve numerical problems using the Julia platform. This book includes recipes on identifying and classifying data science problems, data modelling, data analysis, data manipulation, meta-programming, multidimensional arrays, and parallel computing. By the end of the book, you will acquire the skills to work more effectively with your data.
Julia Cookbook
Credits
www.PacktPub.com
Preface
Free Chapter
Extracting and Handling Data
Metaprogramming
Statistics with Julia
Building Data Science Models
Working with Visualizations
Parallel Computing

## Correlation analysis

Correlation analysis is the process that indicates the similarity and relationship between two random variables. For time series data, correlation analysis would be done between two sets of the datasets. And in non time-series data, correlation analysis would generally be done between two independent variables in the dataset. So, in this recipe, we would look at the correlation analysis of time series (signals).

You have to have the `StatsBase` package ready. This can be done by running:

```using StatsBase
```

### How to do it...

1. Autocovariance is the covariance of a piece of time data with itself at two different time points. This helps in understanding how correlated the time series is, with respect to the time dimension. This metric helps us find trends across the time dimension of a time series across different time points. It can be calculated with default lags using the `autocov()` function:

```autocov(x)
```

The output would look like the following:

2. Now, if we want to set...