Maths for Data Analytics

What is the course about?

Data analytics can be thought of as a process that turns data into information – most of the machinery that accomplishes this is mathematical. This course covers three main types of mathematical machine: descriptive, predictive and transformative.

The descriptive statistics we study include measures of centrality (averages), dispersion (range, standard deviation) and distribution (skewness, kurtosis). Predictive methods are organised around hypothesis-testing and include distribution fitting, regression, analysis of variance and forecasting. As well as making a prediction, we always try to be precise about our confidence in it and to be clear about the assumptions it’s based on.

We also look at some transformational methods that can be used to change our data as a step in the analytical workflow. We particularly focus on sequential data (for example, the kind that comes in a time series) and look at smoothing filters for noise reduction and the Fourier transform, which can extract periodic components from non-periodic data.

What will we cover?

• Mean, standard deviation and higher moments (skewness and kurtosis)
• Linear and polynomial regression
• Hypothesis testing
• Confidence intervals
• Random variables
• The Normal, Exponential and Poisson distributions
• Characteristics of noise and smoothing filters
• Various statistical tests including Student’s t-test, the chi-squared test and the Shapiro-Wilk test for normality
• The general method of Analysis of Variance (“ANOVA”) and why it is necessary
• Bootstrapping
• Kernel Density Estimation
• Bias analysis, with an emphasis on the side-effects of linking datasets together
• A user’s guide to the Fast Fourier Transform (FFT) with applications to time series and forecasting.

What will I achieve?
By the end of this course you should be able to...

• Summarize a dataset using a range of descriptive statistics.
• Formulate hypotheses and test them using a variety of statistical measures
• Infer a population statistic from a sample with an associated confidence level
• Create a statistical model of a phenomenon using techniques such as regression and distribution-fitting
• Extract trend and seasonality from a time series
• Extract noise from data.

What level is the course and do I need any particular skills?

You will need to be comfortable with common manipulations of percentages, fractions and algebra at UK GCSE level. You will not be required to do calculations by hand so this course may be a good opportunity to brush up on those skills.

You don’t need to be familiar with any specific software but it would be helpful to have some prior experience of working with data, whether in a coding setting using R or Python or in a spreadsheet programme such as Excel.

How will I be taught, and will there be any work outside the class?

The classes use a mixture of presentation, worked examples and class discussion.

Are there any other costs? Is there anything I need to bring?

No, all resources will be provided digitally.

When I've finished, what course can I do next?

Please click here to view our Programming and Maths courses.