Share New Research 
I was trying to understand what is being explained at
I understand some parts of the explaination. I am unsure of the Coefficient fo determination section, to me it seems to sugget that between about 5 to 10% of the change in the dependent variable can be explained by the indicator? And I'm not sure if this ‘L’ shaped curve is good or bad? Then they say they make a number of ML models using the sentiment indicator and they seem to not significantly outperform the benchmark in the ‘out-of-sample’ area. So to me it seems the use of the sentiment indicator seems poor and entails quite an effort to get to a final working implementation.
Manav Trivedi
Hello, if you use an ensemble of sentiment analysis and fundamental / quantitative analysis that would yield much better results as sentiment indication on its own will clearly increase the risk of being “wrong” regarding the position you're taking, so you may choose either good financial tool or a heavy machine learning architecture ( from a research paper ) to obtain better results. Thanks.
Louis Szeto
Hi Newoptionz, Manav
Statistically, high R-sqaure value with low p-value would be the most ideal as it means a high correlation between one series to another (r^2) and that relationship is unlikely happened by pure coincident due to sampling (p). So, it would be great if you see scatters concentrated on the bottom-right-hand corner in that plot.
It would be up to you to decide the cutoff score of a reasonable power of explanation for a data source/series since it is very unlikely a single source could dominant in asset pricing (r^2), but alpha/type II error value is usually set as 5% (5% chance of that result is a false positive), drawing a conclusion that a correlation relationship is statistically significant if p < 0.05.
The research notebook is only a very primitive EDA on the dataset. You will need to put on some magic in data cleansing and data processing to transform and exploit the full potential of a data source, and implement your own logic on how to use the generated signals in trading. While our sample is based on one dataset only, you could also combine varies other/custom datasets to yield some amazing result. Hopefully you can share with us if you do😊.
Best
Louis
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.
Newoptionz
The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by QuantConnect. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. QuantConnect makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk, including loss of principal. You should consult with an investment professional before making any investment decisions.
To unlock posting to the community forums please complete at least 30% of Boot Camp.
You can continue your Boot Camp training progress from the terminal. We hope to see you in the community soon!