FinThink

Momentum Materializes Before Share Price Movements

Let's Talk FinTech!

Any time the price of a stock or crypto project goes up, it is because there is interest and following, momentum behind it. FinThink Inc. can now measure the momentum behind market indices and individual stocks and crypto projects.

If there is a lot of interest and momentum behind a stock, its price will either rally or sell off; a stock's share price never goes up without momentum.

Coupled with our NLP NLU AI model, know the direction of the market.

Here is a look at our Momentum Analyzer Algorithm used to gage momentum on market indices and individual securities:

This is FinThink Inc.

finthink.com

Seeing Where the Future Trades

Let's Talk FinTech!

A Neural Network is an artificial intelligence model, a set of algorithms purposely built to closely model the human brain in the way it learns, refines, and cements correct data and forgets incorrect data. It’s purposely designed to recognize patterns. Neural Networks interpret sensory data through machine perception, labeling, and clustering of raw data given to build the model.

Artificial Neural Networks are a large number of processors operating in parallel and arranged in layers. Similar to an optic nerve, the first layer receives the raw input data. Each successive layer then receives the output from its preceding layer rather than from the raw data used in the input layer, similar to how neurons further from the optic nerve receive signals from those closer to it. The last layer, the output layer, produces the output data used for the assigned use case.

A Deep Neural Network (DNN) is a neural network with more than two layers, some of which are hidden layers. DNNs typically have an input, multiple hidden, and output layers.

A Recurring Neural Network (RNN) is a feed-forward neural network composed of highly interconnected processing neurons (processors) working together in layers that can use their internal state (memory) to process sequences of layers to solve a problem, which in this case is forecasting a time series. RNNs can recognize patterns between raw input data and the vector, time series data, given to forecast the given vector.

A Long Short-Term Memory (LSTM), a Recurring Neural Network (RNN) type, is a Deep Learning model used to forecast time-series data. FinThink AI’s LSTM RNN currently forecasts US and Global Equities 30 business days (21 trading days) ahead with a 5%-10% deviation between predicted and actual outcomes with 99.985% accuracy.


This is FinThink Inc.

finthink.com