I like those websites, specially willmott.com. I will check their content, I must get something useful from there. Now that we speak like human beings, the repo I shared on github is just an example for anyone that would like to check my coding style. Obviously the repo that contains algos that I normally use are private. The backtesting folder that you referred to with Technical Analysis is actually to challenge anyone that tells me that TA is useful for anything. I am known in my inner circle of friends as the biggest detractor of TA they can meet. Those coding samples are to prove that any TA indicator can be challenged. When someone tells me that an indicator can make them rich I throw a report to them with a very simple c# script against the first data set I can find. I do not use any TA in my trading, hence the main reason why I managed to be profitable for a while. And when I mean profitable, I mean with a very small account (under 100K) that is not losing during 8 months so far. My algos are based on a very basic risk management process, so your comment about not using rigorous mathematics is on point. I do not use any complex mathematics at all, simply because I don't need them. That doesn't mean that I don't know them and I could use them if required. Changing the topic, I wanted to open another post asking for possible partners to funds my algorithm. I am certainly not going to do that looking at what I see already, but since you have been working in trading firms, what would you require from someone to believe that an algorithm is profitable? Is there any standard format that you would ask for as a proof that I am just not lucky for a while and my algorithms can be trusted? I was just thinking about sharing the Analytics report that my broker is providing but since we can cook anything we want these days with PS that might not be sufficient for someone to believe it.
Unless your strategy is so good that you have to protect it, you can try to put it on sites like quantconnect or other similar sites to seek funding. But if it really is something out of the world, then borrow from friends and family and milk it till you are rich.
I think generally one would expect a transaction report exported from a brokerage account where you implemented this strategy. But I'm just a random guy, why would I want to know? If you believe in your idea,. the thing to do would be to gather money from friends and family, or, take a loan, use something as collateral, sell a kidney, write a business plan, create an LLC and sell membership units, go to the bank and ask for a loan, seek out a vulture capitalists, or one of these so called "angel investor" people . If you form an LLC, you could become a good salesman and sell a few thousand dollars worth of ownership units to up to 50 people or something like that, if you dont know anyone with enough to fund the idea outright. how does it scale over time and across asset classes?risk management is integal but i didnt even go into that at the time because thats a mathematical thing by necessity since option and futures prices are based on expectations, anyhow i digress
I looked at the C++ code, and it wouldn't hurt to have comments. There were several places where it would be better practice to pass function arguments by reference instead of copying. And some methods could be const. For example, https://github.com/oneangrytrader/I....Cpp.Api.Client.DLibrary/include/FileHelper.h
Yeah that header is not one of my worries. You can see exactly what you are saying at https://github.com/oneangrytrader/I...Api.Client.DLibrary/include/IgRestApiClient.h The header you pointed out is used once at the very beginning of any process so I didn't worry too much about it. You have to consider how those methods are used, not every argument should be constant if it is not worth it.