Time Series Analysis with Python

G POORNA PRUDHVI (~poornagurram)


Description:

Time series analysis has been around for centuries helping us to solve from astronomical problems to business problems and advanced scientific research around us now. Time stores precious information, which most machine learning algorithms don’t deal with. But time series analysis, which is a mix of machine learning and statistics helps us to get useful insights. Time series can be applied to various fields like economy forecasting, budgetary analysis, sales forecasting, census analysis and much more. In this workshop, We will look at how to dive deep into time series data and make use of deep learning to make accurate predictions.

Structure of the workshop goes like this:

  • Introduction to Time series analysis: Demystifying terms and basic concepts like Trend, seasonality, white noise, etc.

  • Time Series Exploratory Data Analysis and Data manipulation with pandas: Analyzing for noise, creating features and getting data ready using pandas methods for forecasting.

  • Forecast Time series data with some classical methods: Building baseline with Moving Average and experimenting with commonly used methods for time-series forecasting like Auto-regression, ARMA, ARIMA, GARCH, E-GARCH and comparing them.

  • Introduction to Deep Learning and Time series forecasting: Performing and comparing forecasting with Multilayer Perceptron, Recurrent Neural Network, Long Short Term Memory Neural Network.

  • Financial Time Series data: A deeper look into real-world financial time series data.

  • Forecasting using boosting: We will use the most used boosting technique in Kaggle competitions XGBoost to forecast time series.

Libraries Used:

  • Keras (with Tensorflow backend)
  • matplotlib
  • pandas
  • statsmodels
  • prophet
  • pyflux

Prerequisites:

  • Basics of Python
  • Basics of Time series analysis
  • Basics of Pandas & Deep learning
  • Laptop

The following python packages need to be installed in the laptops of the attendees.

  • Keras (with Tensorflow backend)
  • matplotlib
  • pandas
  • statsmodels
  • prophet pyflux

Speaker Info:

Speaker 1: Gurram Poorna Prudhvi

Prudhvi is working as a machine learning engineer at mroads. He is interested in NLP research, Opensource, Public Speaking, and Python. In his free time he explores and tries to understand different dimensions of life. He is also a core team member of Hyderabad Python Community.

Speaker 2: Ramanathan Ramakrishnamoorthy

Co-Founder, Director & Head of Research & Development at Zentropy Technologies. Before finding Zentropy, Ram worked with a leading hedge fund as a Project Manager responsible for building tools and technologies required by the middle and the back office. He was instrumental in delivering some of the most mission-critical strategic projects that helped in the overall business of the firm.

Having a keen eye for subtle data patterns, Ram also has a great understanding of machine learning and data science domain and applies the right solutions as per the project requirements. He is a specialist in performing exploratory and predictive analysis on Time Series data.

Speaker Links:

Talks:

https://www.youtube.com/watch?v=FbMP187VNTI

https://www.youtube.com/watch?v=RY5QHJow90M

Have done a basic version of the same workshop in scipy India - https://scipy.in/2018#schedule. Please note this was not recorded.

https://github.com/poornagurram/scipy_time_series_python

Id: 1320
Section: Data Science, Machine Learning and AI
Type: Workshop
Target Audience: Intermediate
Last Updated: