You question is completely unclear. You first need to know what you want to study, what is the problem you are trying to solve, and only afterwards it makes sense to think about which numerical method is suitable for you.
I presume that you are looking for a numerical method to successfully trade on stock markets. The first question is who is going to trade?
A small individual investor or somebody with big money. The second question is what is the time horizon for each trade? Do you want to speculate on short time swings of the market or you want to invest longer term? The last question is whether the money is yours and not last or borrowed.
There are many numerical methods applicable to stock markets. Some are in the form of indicators, some are scripts, some are robots. The indicators are mathematical artefacts originating from statistical methods, like moving averages, Bollinger bands, RSI, some are non-stochastic indicators like MACD, OSMA, fractal indicators (Williams) and many many more. Scripts help organise work and facilitate trading. Robots in theory should trade for you with or without your interference. Not a single indicator ensures consistent profits longer term. That is why on top of that we have trading strategies and money management rules which are top important for any serious investor.
You can take a look at my coauthored papers on "Mathematical Statistical Pricing in (Electronic) Stock Markets of Emerging Markets" and of Electronic Banking Sector on www.researchgate.net/Profile/Soumitra_Mallick for some Gauss Markov reduction methods with our developed Genetic Algorithms and Systems Classification and Systems Integration of Mathematical-Statistical spacextime data experiments using electronically generated hence genetic cybernetic data. It is a String Theory Method for solving Pvs. NP type problems (Mallick, Hamburger & Mallick (2016, 2017)) and instead of the Newton-Raphson or Trapezoidal Geometric Methods uses Mathematical Physics methods with Vector Diffeotopic Embedding String Matching Field Genetic Algorithms with randomised protocols (Mallick (1993, 2012, 2013, 2014, 2015, 2018)). It is not my intention to make research difficult but Mathematical Physics is not easy and both new Laws & Methods are necessary for analysing String Matching Fields with Genetic AGGNNNetworks for Distributed Computing Systems with Entropy. You can take a look at them if you wish to.
Soumitra K. Mallick
Soumitra K. Mallick, Nick Hamburger, Sandipan Mallick of RHMHM School
I have emailed you one of the most recent publications which contains the references to some of the other papers. We hope you find them creating a New String Field Theory and Method by applying a new calculus (Mallick, Hamburger, & Mallick (2016)) with AGGNNNetworks.
First of all i would like to inform you that their are N number of technical indicators which can be used and a lot of research has been done on the profitability of those indicators. But it may be possible that if certain strategy or indicator is profitable in a country, but it may not be profitable in some other country. I have done a detailed research on profitability of Moving Averages ( Five variations of MA, including SMS, EMA, TEMA, DEMA) and almost on all oscillators which are commonly used by traders including RSI, Stochastic, CCI, etc.
You can reach my following papers to get detail insight.
Article The Profitability of Five Popular Variations of Moving Avera...
Article Technical Analysis of Indian Financial Market with the Help ...
Article Profitability of Oscillators Used in Technical Analysis for ...
Feel free to contact me if you need any kind of collaboration or help with your research.