Tasks and priorities.

[nikolas-joyce]

I am still thinking about having you duplicate some of the optunity notebooks and the active member selection process from V1. 1)At this point i think it is going to be unavoidable but i think i will ask you to do as much as you can in 5-6 hours for that project.

I am still trying to get some V2 parts into a sample campaign for testing.

2)could you remind me will an alpha developed in V2 work in the campaign infrastructure from V1 and therefore be accessible for real-time trading if needed?

3)If everything which is developed in V2 is compatible then the next hurtle on our side is the full product catalogue migration of the data to the V2 Mongo? I think Steve confirmed he knows who to script this transition.

4)The last project i was hoping to get to was the development of an alpha off of this ML work Dmitry had done here https://10.0.1.2:8888/notebooks/development/ML%2C%20position%20sizing%20model.ipynb.

5) I came across a potential client would could definitely make use of our infrastructure. i am hoping to develop a consulting agreement so i will have adequate budget to keep you occupied. I will keep you posted.

on that note i was thinking how we could deploy and scale our infrastructure to multiple independent users and i came across concept of the droplet.

https://www.digitalocean.com/community/tutorials/how-to-create-your-first-digitalocean-droplet-virtual-server https://www.youtube.com/watch?v=yX3I8qumJes

I was thinking we could push some limited version of our git repository to the droplet and allow others to access our infrastructure. any thoughts on the viability of this as a concept?

[alexveden]

2)could you remind me will an alpha developed in V2 work in the campaign infrastructure from V1 and therefore be accessible for real-time trading if needed?

yes, correct, I've built campaign bridge between frameworks. So all campaign positions of v2 alphas will be available for execution software. You just need to feed the real time data for v2 alphas and indexes.

3)If everything which is developed in V2 is compatible then the next hurtle on our side is the full product catalogue migration of the data to the V2 Mongo? I think Steve confirmed he knows who to script this transition.

I'm not sure that I'm clearly understand the question. But I believe that I've implement all nuances related to all products which we have faced while used v1.

4)The last project i was hoping to get to was the development of an alpha off of this ML work Dmitry had done here

As I can seen this is pretty rough proof of the concept for ML application for position sizing, because Dmitry has used limited features set (jut few TA indicators). But I want to believe that I made everything to add support for ML models for alphas and portfolios.

I came across a potential client would could definitely make use of our infrastructure. i am hoping to develop a consulting agreement so i will have adequate budget to keep you occupied. I will keep you posted.

Sounds great.

on that note i was thinking how we could deploy and scale our infrastructure to multiple independent users and i came across concept of the droplet.

Droplet is ok, but like an other VM it has to be reconfigured each time it has been created. I didn't dig deeply, but I think this is doable.

I was thinking we could push some limited version of our git repository to the droplet and allow others to access our infrastructure.

I have heard about Docker containers which helpful with code deployment, probably it could be better solution. I'm not sure yet, but it could be possible to use docker containers within single high-end dedicated server and sell subscription to micro services of docker containers. I mean you could get extra revenue by selling subscription, instead of using digital ocean services.

Useful links: https://www.docker.com/what-docker https://www.docker.com/use-cases

[nikolas-joyce]

Relating to the task i had in mind after you finish the hedge ratio/risk parity spread creation i thought i would outline the notebooks i use here. I believe you thought it would be necessary to the implement V1 selection algorithm to get similar result in V2. Also i wanted to have V1 style flexibility to setting different objective function and swarm visibility that i could not find in the current V2 base and genetic opt.

https://10.0.1.2:8888/notebooks/strategies/Strategy_DSP_LPBP_Combination_StrategyModule_new%20%2B%20Optunity%20smart%20optimization.ipynb

this notebook, which Dmitry build for the various versions of the production alphas, has a few useful features that i wanted to highlight.

1) the ability to optimize based on a selection of different objective functions. 2) ability to alter the size of the in and out of sample data usage. 3) functionally formatted solutions that can easily be copied into another notebook for deployment.( this feature may be redundant with your v2 alpha.save function)

4) the ability to examine visually the resulting sub swarm and picked equity curve.

[nikolas-joyce]

BTW is it possible to view the existing V2 alphas outcomes in the the way outlined above or is the concept of a swarm and a calibration not compatible with V2 alphas?

[nikolas-joyce]

on this 4)The last project i was hoping to get to was the development of an alpha off of this ML work Dmitry had done here https://10.0.1.2:8888/notebooks/development/ML%2C%20position%20sizing%20model.ipynb.

my thinking was we would develop it as a stand alone alpha type that would require a minimum of 3-5 units of the options to function. if 5 was the strategic quantity and the ML says the size should be .5 then we would run 3 etc. is this concept clear?