https://github.com/noqcks/forex_algo

Just fooling around with forex trading
https://github.com/noqcks/forex_algo

Last synced: 20 days ago
JSON representation

Just fooling around with forex trading

• Host: GitHub
• URL: https://github.com/noqcks/forex_algo
• Owner: noqcks
• Created: 2017-04-13T15:43:39.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2017-04-18T13:39:10.000Z (over 7 years ago)
• Last Synced: 2024-10-11T16:12:54.711Z (3 months ago)
• Language: Python
• Size: 226 KB
• Stars: 3
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Sentiment Based FOREX Algo

This is still technically a work in progress until I can find enough headline

data to validate my algo.

This will try predict FOREX market movements from headline data.

1. gets headline data from eventregistry

2. gets forex data from dukascopy

3. transforms headline data and associates it with forex data so that for every

2H period a string of headline information will have a P (pos) or N (neg) rating

based on forex data. There is a 1H offset between the forex data and headline data

so that we give the market time to converge.

4. this is where we clean the headline data and transform it. We take out low

information words and symbols, get word hypernyms, and then apply a sentiment score

to each word. We transform the hypernyms into a tfidf vector and then multiply our

tfidf scores by the sentiment sum scores we got earlier.

 The sentiment sum score is defined as:

 ```

sentiment_sum_score = word_negative_sentiment + word_positive_sentiment

```

5. this is where we apply a machine learning classification model (SVM) that will

try predict price movements from the headline vectors.

## Config

You can alter config settings in config.ini

```

start_date: the start date you want to analyze

end_date: the end date you want to analyze

er_api_key: the API key for http://eventregistry.org/

currency_pair: the currency pair you would like to try predict

```

## Running

```

./run.sh

```

## Requirements

```

pip install -r requirements.txt

```