Statistical analysis and visualization of state transition phenomena

Share Market Prediction App using Markov Chains Model

Continuous Time Markov Chain

Application of Markov Chain in Finance

Scripts supporting the Open Risk Academy course Analysis of Credit Migration using Python TransitionMatrix

This application makes predictions by multiplying a probability vector with a transition matrix multiple times (n steps - user defined). On each step the values from the resulting probability vectors are plotted on a chart. The resulting curves on the chart indicate the behavior of the system over a number of steps.

Predictions with Markov Chains is a JS application that multiplies a probability vector with a transition matrix multiple times (n steps - user defined). On each step, the values from the resulting probability vectors are plotted on a chart. The resulting curves on the chart indicate the behavior of the system over n steps.

A Markov-chain based supermarket simulation.

The Markov Chains - Simulation framework is a Markov Chain Generator that uses probability values from a transition matrix to generate strings. At each step the new string is analyzed and the letter frequencies are computed. These frequencies are displayed as signals on a graph at each step in order to capture the overall behavior of the MCG.

The current JS application is a detector that uses observation sequences to construct the transition matrices for two models, which are merged into a single log-likelihood matrix (LLM). A scanner can use this LLM to search for regions of interest inside a longer sequence called z (the target).

This application uses a transition matrix to make predictions by using a Markov chain. For exemplification, the values from the transition matrix represent the transition probabilities between two states found in a sequence of observations.

The transition matrix of a Markov chain is a square matrix that describes the probability of transitioning from one state to another.

Experimenting with the transition state matrix approach to credit default modeling.

WeatherChance is an open-source application that can predict whether the tomorrows weather of particular queried location/city will be good or bad. Good weather is essentially defined as sunny and less cloudly and bad weather is defined as rainy, snowy etc.

Modeling and visualization of the movement of supermarket visitors based on real customer data.

Simulates the movement of players around the board for a game of US Standard 2008 Edition Monopoly, using a Markov process, in order to model the likelihood of landing on each tile.

NPM package to easily create and use Markov chains

Reinforcement Learning Using Q-learning, Double Q-learning, and Dyna-Q.

A Monte Carlo simulation representing the daily behaviour of customers inside a fictional supermarket. Featuring a colourful and clear visualisation interface.

Library to find the Probability Estimation of Navigation Paths and their Pattern Prediction.