From now on, I'm going to build an executable solution to manage investment or speculation activities in order to earn more with less effort. I plan to go it through with a bottom-up approach. This is my initial imagination:

First of all, I'll code a prediction function to take wild guess with numbers. It solves problems such as:
What number is most likely to be in the question mark? {1,2,2,3,4,5,4,5,6,5,6,5,6,6,?}
Obviously it's the most important and complex part of this solution, it's inevitable to redesign this function again and again, so, it's acceptable to start with a wild guess at this stage.

Then, real data must be imported. Real data play two roles in this solution. One is a material to be taken in back-test by which the prediction function is assessed and improved. The other one is a prerequisite of various analyses which serves portfolio management.
Although I exclude the content of media press and social networks from real data, the integration of real data is extremely difficult, I plan to restrain the usage of real data otherwise it will cost me decades.

As mentioned, the content of media press and social networks could offer some basis of prediction. Technologies such as crawler and natural language processing are necessary.

Till then, the embryo of this solution is ready. Possible improvements could be:

  • Optimize or rewrite the defined analytics modules.
  • Implement machine learning technology to make better use of data and information.
  • Create a module to monitor and log performance.

Now, let's set aside all the complexity and focus on this question:
What number is most likely to be in the question mark? {1,2,2,3,4,5,4,5,6,5,6,5,6,6,?}

标签: none

Add a new comment.

  • 上一篇: 没有了
  • 下一篇: 没有了