Regression and Time Series Analysis

Linear regression, nonlinear regression, auto-regressive models, time series analysis, regularisation and more.

Next event: 24 Jun - 25 Jun 2017 in London


Level

intermediate

LONDON

CAMBRIDGE

OXFORD

Duration

two days Course



What you will learn

You will learn the skills you need to develop and evaluate regression models that allow you to make quantitative predictions.

  • Multivariate linear regression
  • Ridge and LASSO regularisation
  • Logistic regression and generalised linear models
  • Time series analysis: AR, MA, ARMA and ARIMA models

Languages and libraries

  • Python 3
  • Numpy and Pandas for data manipulation
  • scikit-learn and statsmodel for linear and time series models
  • matplotlib for visualisation

Progression paths

Learn state-of-the art machine learning techniques at our Machine Learning Techniques using Python bootcamp.

Acquire specialised Natural Language Processing skills at our Text Mining and Natural Language Processing with Python bootcamp.


Prerequisites

Audience: Those who wish to take their data science skills further and learn state-of-the-art techniques in this constantly evolving field.

Prerequisites: Good knowledge of python, basic understanding of machine learning practice (as taught in Introduction to Data Science)

Day 1

Regression analysis

Session 1

Linear regression

  • Multivariate regression using SKLearn
  • Outliers, leverage, analysis of diagnostics
  • Data transformations

Session 2

Regularisation: Ridge and LASSO

  • The Ridge regression
  • The LASSO regression
  • Applications and comparisons

Session 3

Generalised linear models

  • The Logistic regression
  • Use of regularisation
  • Applications in industry

Evening

Social

  • Drinks with fellow participants and lecturers

Day 2

Time series analysis

Session 1

Handling time series

  • Manipulating time series with Pandas
  • Handling trends and seasonalities in time series: understanding lagging
  • Autocorrelation function and moving average (MA)

Session 2

Time series models

  • Building an autoregression (AR)
  • Adding structure in a time series model: towards ARMA and ARIMA models
  • Modeling real-world time series: how to avoid pitfalls

Continuous learning project

Our continuous learning project comprises a real-world problem and data set to complete in your own time, and practice using the course material and techniques covered during the bootcamp. The package includes model notebook answers, with a detailed explanation of the solution and problem-solving process.

Price: £100 extra

Highlights

Check out video highlights, photos and interviews from our previous bootcamps.


In-house Training

Get in touch to discuss your requirements by emailing contact@cambridgespark.com or by completing our contact form.

We can deliver this course as a private training at your office during week days.

We can also design a bespoke curriculum matching your specific training objectives.

Book your ticket

Event:
24–25 Jun, London
Location:
THECUBE - Studio 5 , 155 COMMERCIAL STREET, E1 6BJ, London (London)
Ticket:
Ticket includes course materials and code resources.