Karpet is a tiny library with just a few dependencies for fetching coins/tokens metrics data the internet.

It can provide following data:

- coin/token historical price data (no limits)
- google trends for the given list of keywords (longer period than official API)
- twitter scraping for the given keywords (no limits)
- much more info about crypto coins/tokens (no rate limits)

— Permalink

You will learn in this post how to:

- decompose double-seasonal time series
- detrend time series
- model and forecast double-seasonal time series with trend
- use two types of simple regression trees
- set important hyperparameters related to regression tree

This web site contains notes and materials for an advanced elective course on statistical forecasting that is taught at the Fuqua School of Business, Duke University. It covers linear regression and time series forecasting models as well as general principles of thoughtful data analysis.

The time series material is illustrated with output produced by Statgraphics, a statistical software package that is highly interactive and has good features for testing and comparing models, including a parallel-model forecasting procedure that I designed many years ago.

The material on multivariate data analysis and linear regression is illustrated with output produced by RegressIt, a free Excel add-in which I also designed. However, these notes are platform-independent. Any statistical software package ought to provide the analytical capabilities needed for the various topics covered here.

— Permalink

This article focuses on using a Deep LSTM Neural Network architecture to provide multidimensional time series forecasting using Keras and Tensorflow - specifically on stock market datasets to provide momentum indicators of stock price.

The following article sections will briefly touch on LSTM neuron cells, give a toy example of predicting a sine wave then walk through the application to a stochastic time series. The article assumes a basic working knowledge of simple deep neural networks.

]]>When you are presented with a new time series forecasting problem, there are many things to consider.

The choice that you make directly impacts each step of the project from the design of a test harness to evaluate forecast models to the fundamental difficulty of the forecast problem that you are working on.

It is possible to very quickly narrow down the options by working through a series of questions about your time series forecasting problem. By considering a few themes and questions within each theme, you narrow down the type of problem, test harness, and even choice of algorithms for your project.

In this post, you will discover a framework that you can use to quickly understand and frame your time series forecasting problem.

— Permalink