Analyze data set to find top factors that drive outcomes
This project was successfully completed by Kuntalbakuli for $111 USD in 2 days.Get free quotes for a project like this
Project Budget$30 - $250 USD
Completed In2 days
Please read fully before bidding
The attached spreadsheet contains a series of "real world" results for 4 sets of currency pairs based on formulae designed to predict the results or "expirys". Green indicates an accurate prediction and red indicates an inaccurate prediction. The summary tab contains the overall summary of results for each pair by prediction type.
Key terms and definitions (use the following chart to calculate accuracy trends)
E = expiry (these occur at quarter-hour intervals after the Alert Price/Time)
Cluster 1 = E 1-4
Cluster 2 = E 4-6
Cluster 3 = E 6-8
Cluster 2,3 = E 4-8
Cluster 4 = E 1-8 (all 8 expirys)
Note: note expirys within clusters need to be analyzed separately. For example, do not group results for Cluster 1 with the results Clusters 2,3. These are separated in the spreadsheet.
1) analyze the results and for each currency pair, let me know the best time of day, day of the week, or price leads to the best results. The goal is to provide a set of rules that result in >95% accuracy on the expirys in the cluster. Please note, I am only interested in the rules for clusters where all of the expirys in the cluster were green.
2) Overall, for each pair, I would like you to analyze and send me 2-3 simple rules that says, for example, if you place a bet between 2-4GMT on Tuesdays the results for USDJPY will almost always move as predicted (e.g., the full cluster or most of the expirys in the predicted cluster are green) or Sundays between 15-23GMT the EURUSD predictions are almost always accurate.
Overall, the goal is to find the rules that result in >95% accuracy on the expires (all expirys in the cluster are green). Note, the current accuracies range between 46-84%.
Feel free to send me questions.
Looking to make some money?
- Set your budget and the timeframe
- Outline your proposal
- Get paid for your work
Hire Freelancers who also bid on this project
Looking for work?
Work on projects like this and make money from home!Sign Up Now
- The New York Times
- Wall Street Journal
- Times Online